Role Of Data Science In Making Data-Driven Decisions In App Development Companies <\/p>\n
2<\/p>\n
Role Of Data Science In Making Data-Driven Decisions In App Development Companies <\/p>\n
2<\/p>\n
THESIS PROPOSAL TEMPLATE<\/p>\n
Title: ROLE OF DATA SCIENCE IN MAKING DATA-DRIVEN DECISIONS IN APP DEVELOPMENT COMPANIES<\/p>\n
Prepared<\/p>\n
by<\/p>\n
Rajyalakshmi Kommineni<\/p>\n
University of the Cumberland\u2019s<\/p>\n
Ph.D.<\/p>\n
DATE:11\/26\/2021<\/p>\n
[Cover Page]<\/p>\n
Title: ROLE OF DATA SCIENCE IN MAKING DATA-DRIVEN DECISIONS IN APP DEVELOPMENT COMPANIES<\/p>\n
A proposal submitted in partial fulfillment of the requirements for the Ph.D. In Information Technology<\/p>\n
Name: Rajyalakshmi Kommineni<\/p>\n
Proposed Supervisor<\/p>\n
Dr. Janita Haastrup \u2013 (Ph.D.)<\/p>\n
Date of Submission<\/p>\n
11\/26\/2021<\/p>\n
ROLE OF DATA SCIENCE IN MAKING DATA-DRIVEN DECISIONS IN APP DEVELOPMENT COMPANIES<\/p>\n
A Doctoral Dissertation Research<\/p>\n
Submitted to the<\/p>\n
University of the Cumberland\u2019s<\/p>\n
In partial Fulfillment of <\/p>\n
The Requirements for the Degree of<\/p>\n
Doctor of Education<\/p>\n
By<\/p>\n
Rajyalakshmi Kommineni<\/p>\n
November 2021<\/p>\n
ROLE OF DATA SCIENCE IN MAKING DATA-DRIVEN DECISIONS IN APP DEVELOPMENT COMPANIES<\/p>\n
Copyright \u00a92021<\/p>\n
Rajyalakshmi Kommineni<\/p>\n
All rights reserved<\/p>\n
Approval for Recommendation<\/p>\n
This dissertation is approved for recommendation by the faculty and administration of the University of the Cumberlands. <\/p>\n
Dissertation Chair:<\/p>\n
Dissertation Evaluators:<\/p>\n
Acknowledgments<\/p>\n
There are many to whom a debt of gratitude is owed for their assistance in conducting this research\u2026. (It is appropriate to thank key faculty, friends, and family members, as well as ministers and God. It is advisable to limit the comments to one page)<\/p>\n
Abstract <\/p>\n
This study examined the differences\u2026\u2026\u2026\u2026\u2026\u2026<\/p>\n
Table of Contents<\/p>\n
Chapter One: Introduction<\/p>\n
Overview\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..\u2026…………….9<\/p>\n
Background and Problem Statement\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026…\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026\u2026\u2026..\u202611<\/p>\n
Purpose of the Study\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026\u2026…\u2026..16<\/p>\n
Significance of the Study\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026…\u2026…18<\/p>\n
Research Questions\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..\u2026.24<\/p>\n
Theoretical Framework\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026\u2026..\u2026\u202625<\/p>\n
Limitations of the Study Assumptions\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026…27<\/p>\n
Definitions\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.\u2026.\u202630<\/p>\n
Chapter 2 Literature Review<\/p>\n
Introduction\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.45<\/p>\n
Literature Review\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.46 <\/p>\n
What is data-driven decision making\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202647 <\/p>\n
Benefits of Data-driven decision making\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..48 <\/p>\n
Data science and its relationship to big data and data-driven decision making\u2026\u2026\u2026\u2026\u2026\u2026\u2026.49<\/p>\n
The emerging role of data scientists on software development teams\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..50 <\/p>\n
Data-intensive applications, challenges, techniques and technologies\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..51 <\/p>\n
Mobile networks and applications\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.51<\/p>\n
Data-driven dashboards for transparent and accountable decision-making in smart cities\u2026\u2026..52 <\/p>\n
Data science life-cycle\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026…53 <\/p>\n
Designing of data-driven apps\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..55 <\/p>\n
Data-intensive applications, challenges, approaches, and innovations\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.56 <\/p>\n
Big data methods and technologies\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202657 <\/p>\n
Big Data Techniques\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.58 <\/p>\n
Big data analytics and application for logistics and supply chain management\u2026\u2026\u2026\u2026\u2026\u2026\u202660 <\/p>\n
Data-driven smart manufacturing\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..61 <\/p>\n
Data science, predictive analytics, and big data\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202662 <\/p>\n
Real-time big data processing\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202663 <\/p>\n
Web-analytics, social media, and mobile apps\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..64 <\/p>\n
Role of data scientists on application development teams\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202667 <\/p>\n
Public policy considerations for data-driven innovation\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202668 <\/p>\n
Applications, prospects and challenges\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.70 <\/p>\n
Creating a data-driven organization\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.70 <\/p>\n
Toward data-driven requirements engineering\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..71 <\/p>\n
The opportunity and challenge for IS research\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..72 <\/p>\n
The role of data-driven e-government in realizing the sustainable development goals in developing economies\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202673 <\/p>\n
Big data: the management revolution\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202678 <\/p>\n
A taxonomy of data-driven business models used by start-up firms\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.78 <\/p>\n
Management Review\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.79 <\/p>\n
Data mining with big data\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..79 <\/p>\n
Harness the power of big data\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202680 <\/p>\n
Data privacy and security\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026…80 <\/p>\n
A data-driven organization\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202681 <\/p>\n
Challenges in mobile app development\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.82 <\/p>\n
Conclusion\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.85 <\/p>\n
Chapter 3: Procedure and Methodology<\/p>\n
Introduction\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202687<\/p>\n
Research Paradigm (quantitative)\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..89<\/p>\n
Research Project Design\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u202695<\/p>\n
Sampling Procedures\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026.99<\/p>\n
Statistical Tests\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026102<\/p>\n
Summary\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026..\u2026\u2026102<\/p>\n
Index\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026107<\/p>\n
Chapter One<\/p>\n
Overview<\/p>\n
Contemporary organizations depend on effective data management of assets. Data analytics and management have developed into crucial success agents for companies over the last few decades. The various aspects involved in changing a company to a data-driven one should be studied. Many angles should be considered when changing to a data-driven organization. The literature looks at major factors and corresponding needs of data-driven companies to create a consensual description that could be used in research. The study is based on organizational design and indicates deductively generated findings gathered from the literature assessment. Significant contributions to the paper are the common comprehension and strategies of significant factors in data-driven organizations. <\/p>\n
The constant penetration and development of digital innovation have resulted in disruptive shifts in society and the economy. Digital innovation allows for new methods of using data to increase business performance and find new arrangements for efficient data-driven models. What makes data different from other resources is the ability to allow recurring services while incurring an almost zero marginal cost and provide plenty of utilization decisions autonomous from distinct devices (Veit et al, 2014). Companies must incorporate analytical technologies like prescriptive or predictive assessments to apply data’s potential in unmasking knowledge and undiscovered opportunities (Berndtsson et al. 2018). Articles use the phrase ‘data-driven to show data’s purpose as a key factor (Hartmann et al, 2016). In all the studies conducted in this research, various literature is examining the phrase data-driven. Jurgen (1983), while looking at the automation processes driven by data in manufacturing industries, stated that it is the, more innovative application of data that results in accomplishing significant cost cuts. Now, people understand that the phrase data-driven is appropriate for all organizations and society. Predictions show that the quantity of data stored globally will grow five times (Reinsel et al., 2018). Consequently, looking at data’s economic value is a topic that must be discussed. Companies need to align with the unpredictable and rapid change procedures and incorporate data into new and existing methods. Since there still lacks a good definition for such organizations; organizations should continue looking for key factors and extracting conceptual needs. These organizations can be described as data-driven organizations. <\/p>\n
Decisions are fundamental when it comes to competitiveness and performance within an organization. Failures and accomplishments all through humanity have been associated with one, fate-altering decision. Also, massive weight has been bestowed on the decision-makers to ensure sure that the most effective attainable decision is made timely (Bartkus et al., 2018). Beginning with how they were made, the traits that describe the decision-makers, the processes involved, and whether they can be achieved have always attracted researchers. This has resulted in a surplus of studies on decision theory and decision-making in recent decades. Based on the concentration of intellectual subjects like sociology, mathematics, political science, and economy, various theorists have wondered what decisions say about different entities, their values, how they can be interpreted and improved (Gigerenzer &\u00a0Gaissmaier,\u00a02015). This resulted in studies on decision-making and organizational behavior, uncertainty, rationality, vulnerability, optimization, and support tools (Simon, 1959). This together with an improved comprehension of human attributes and improvements in technology that simulate and sustain cognitive procedures, decision-making has improved in various situations. <\/p>\n
Background and Problem Statement<\/p>\n
In the mid-twentieth century, studies began detailing decision systems and this included the decisions people make alongside machines and software that had enhanced predictive abilities (Simon, 1977). This increased the interest in decision systems in terms of systems, people, processes, and data for sustaining decision procedures and making autonomous decisions (Power, Heavin, et al., 2019). <\/p>\n
Furthermore, with improvements in machine learning, data science, analytics, data-instigated decision-making based on findings and evidence indicated through analytics, has increased in popularity (Intezari &\u00a0Gressel, 2017). This has made data-driven decision-making to be considered as a way of providing quality decisions based on facts since it integrates human experience with data analytics to create room for rational decisions thus improving outcomes (Janssen et al., 2017; Power, 2016; Provost & Fawcett, 2013).<\/p>\n
The potential of tapping big data and the increasing focus on big data analytics has also increased the focus by assuring improved decision-making by applying the abilities possessed by machines and humans and their unified integration (P. Grover & Kar, 2017; Gupta et al., 2018). With the surge in options and volume of data that can be integrated and assessed like video applications, various analytical methods in different industrial sectors have been the focus of research with the ultimate objective being the improvement of decision-making. This has allowed for decision-making processes that involve a huge amount of collaboration as a result of the influx of information arising from various information sources and the use of analytics (Gigerenzer &\u00a0Gaissmaier,\u00a02011).<\/p>\n
Also, the flourishing conjunction of AI-supported systems with human decision-makers in companies has propelled the interest in increasing their abilities and intelligence, resulting in improved data assessments and consequently improving and sustaining decision-making (P. Grover et al., 2020; Kotsiantis et al., 2006). The integration has resulted in several intelligence dimensions and a variety of applications with various complexities with the objective of creating the best integrating machine and human abilities for improved data-instigated decision-making (Trunk et al., 2020). Thus from bolstering our natural decision-making abilities to completely automating decision procedures, the dependence on technology as a significant part of organizational decision-making is increasing. Though despite the increase in the volume of data, insights, and instruments, there is a gap in harnessing the potential of current innovation, particularly without explicitly defined procedures and guidelines, thus the need for more research. <\/p>\n
There is always a reason why websites and applications like YouTube and Amazon show a person the content they are most likely to interact with. This can be attributed to Machine Learning. This allows organizations to offer customized content and involve an increasing number of users by using machine learning and AI. Machine learning applications are increasingly getting embedded into people’s daily lives as innovation develops towards offering effective solutions that are mobile-centric (Elragal &\u00a0Klischewski, 2017). <\/p>\n
Big data has developed over the past few years as a new concept that offers numerous opportunities and data to enhance studies and decision-support mechanisms with unparalleled value for innovations in sciences, businesses, and engineering (Manyika et al., 2011).\u00a0 Concurrently, big data has its complications when it comes to storage, movement, processes, extraction, and the purpose of data. Breakthroughs like cloud computing offer significant support to deal with the challenges of common computing resources like networking, computing, and storage software (McAfee et al., 2012). Such breakthroughs have led to big data developments. The paper will also look at two perspectives, cloud computing, and big data and assess the gains and risks of applying cloud computing in handling big data in various science disciplines. Based on the introduction, hurdles, sources, research opportunities, and innovation status, the following conclusions are made. Big data and cloud computing encourage application developments and scientific discoveries. Cloud computing is an important factor in solving the storage problem associated with big data (Mikalef et al., 2018). Different app domains and spatiotemporal reasoning help in improving cloud technology and associated innovation with new requirements. Fundamental spatiotemporal concepts of geospatial disciplines and big data provide the resource for locating theoretical and technical solutions to improve big data processing and cloud computing, and the availability of big data and its processing abilities are social risks that have a geospatial weight. The literature will indicate future innovations and study programs for cloud computing that sustain the change in volume, variety, velocity, and veracity into big data values for digital applications and learning. \u2003<\/p>\n
AlgorithmWatch published 2019 an assessment on automated decision-making in Europe. The comprehensive assessment showcased various cases in which companies and governments in the continent had implemented automated decision-making or analyzed the effects of data-instigated decision-making (Kearny et al., 2016). Also, the report offers an understanding of the significance of data-instigated decision-making and its effects and why more research should touch on future decisions. Data-instigated decisions are still experiencing difficulties, despite the vast possibilities of innovation, AI, and automation and why they should be researched and discussed. An example, according to Ioannidis et al. (2020), showed how decisions on forecasting during the recent pandemic were ineffective due to inadequate data input, poor previous evidence, wrong modeling assumptions, a lack of accountability, and the fact that various dimensions were left out when coming up with results (Diakopoulos, 2016).<\/p>\n
Bean and Davenport (2019) also stated that organizations are failing in their quest to be data-driven when one looks at the surprising findings indicated by various study groups (NewVantage Partners, 2019). The results indicate that there is still a wide gap in creating a data-driven culture in which data is considered an essential business asset. This is why it is provided with a lot of investments, attention, and resources. <\/p>\n
Another study by Frisk and Bannister (2017) indicated that even though the skillful application of big data and data analytics can significantly improve an organization’s performance, leaders should change their decision-making aspects to accomplish such developments. The research that concentrated on various fire and rescue service entities in Europe indicated a structural issue in how decisions were motivated. Even though they had a robust technological background, the decisions made were made in data silos since the users were not actively involved in making them. The IT departments were also not seen as strategic resources. Additionally, too much concentration was given to innovation when purchasing expensive ICT systems, but this would reduce when it came to an understanding of how they affected decision-making and their role in the organization. The integration of analytics and big data needs organizations to move towards a considerate and systematic way of data-instigated decision-making. <\/p>\n
Also, poor decisions call for the understanding of past endeavors to lead decision-makers away from practices prone to failure (Nutt, 2010). Recent studies from MIT Sloan indicate that organizational learning calls for machines and humans to work together and learn from each other, thus the creation of collective knowledge between AI and humans based on human experience and digital data to apply innovation, data, and algorithms. However, this calls for major changes, effort, and consideration of machine and human interaction regarding the situation and decision type (Ransbotham et al., 2020).<\/p>\n
Recent studies have indicated the significance of analytics, big data, and AI and their effects on decision factors like efficiency, quality, and success. Though such research, when inevitably dealing with significant challenges and opportunities, has failed to provide a general concept in data-instigated decision-making that includes all these parts. Also, even though the methods, approaches, and theories involved in decision-making have managed to withstand various tests involving applicability and time, each era calls for the creation of additional methods to existing theories to maintain environmental shifts and advances in innovation (Oliveira, 2019). The existing information system studies have only managed to turn square pegs into round holes by being forced to work with current theories and forcing them to discuss what is unachievable with the existing advances in innovation (Gregor, 2006).\u00a0<\/p>\n
Even though different explanations and solutions have been investigated for various phenomena and concepts in data-instigated decision-making, a concept that relies on the scientifically valid concepts of classical decision theory while looking at the correlation between the various elements has not been proposed (Peterson, 2011). The research will address the perceptive lack of concepts that accommodate the contemporary factors of data-instigated decision-making. <\/p>\n
Previous studies on Big Data processing concentrated on stream-based and distributed procedures (Zikopoulos and Eaton 2011). Even though the concept of cloud computing was formed before big data, it is considered an innovative paradigm for achieving computation with attributes such as pooled resources, self-service, elasticity, on-demand access, and pay-as-you-go. The traits allow cloud services to be considered as infrastructure, platform, and software services (Mell and Grance 2011). Having a role in defining the opportunities of different disciplines, cloud computing and big data have resulted in improved solutions for various problems in various disciplines like business, social science, industry, and astronomy. <\/p>\n
Also, the increasing need for precision, big data, and improved resolutions have encouraged the development of cloud computing and related innovation. Cloud computing offers innovative methods of doing business and supports creativity (Zhang and Xu 2013). Bringing together the economy, big data, and computing of services and goods have encouraged various conversations on IT-related concepts since they are part of everyday purchasing procedures (Huang and Rust 2013). There are proposals that the big data entities that have 5V features and issues will continue encouraging the rapid development of associated cloud computing innovation in various perspectives.<\/p>\n
Purpose of the study<\/p>\n
The research tries to determine significant factors of data-driven businesses and their conceptual needs. As indicated by Webster and Watson (2002), a literature assessment was conducted. To answer the research questions, the study uses a joint structure showcasing significant factors of data-driven organizations and their attributes. Conceptual needs derived from various studies were constructed and based on organizational design theory to determine the research questions. The research assesses and structures related study concepts and is followed by many implications. <\/p>\n
To begin, the research proposes a platform to create an understanding concerning the meaning of a data-driven organization and the main factors it contains. Each significant factor indicates an opportunity for thorough study projects that are valid, correlated, and contextualized via the structure. Even though the structure is high-level, it allows one to look at these organizations in a simple way. This is a limitation due to various detail abstractions and a contribution due to its harmonizing trait. The findings can guide leaders on how to identify appropriate areas that need action in creating data-driven organizations while at an optimum level when considering practical effects. Of significance is the fact that the research can act as a checklist for important aspects of the transformation procedures. Uniquely, the structure addresses managerial positions that concentrate on data, like the Chief Data Officer position to guide the transformation procedures effectively. Also, the research provides a descriptive value that indicates the attributes of a data-driven organization and provides suggestions on the areas that should be considered with the utmost care. <\/p>\n
Also, the research intends to answer the identified research questions by looking at various kinds of literature to create a decision theory that accounts for the abilities of data-instigated decision-making by incorporating the decision-making factors with current advances in data analytics and big data. The theory seeks to act as an epistemological foundation that supports the engagements of data-instigated decision-making by offering explanations beyond the theory, thus allowing for future studies in this field. To circumvent common gap-spotting assessments and carry out innovative research, the concept is explained using a study known as problematization as a methodology by Alvesson and Sandberg (2013). This calls for the application of various methodological concepts, including recognizing a field of literature, recognizing and explaining assumptions that characterize the field, assessing the assumptions, creating an alternative, looking at how it relates to the audience, and assessing the alternative assumptions. Nothing discussed in the research falls outside the domain of data science. The methods all apply data, statistics, and programming to gather significant conclusions on the world. The phrase data science is valuable since it correlates with a more comprehensive set of approaches and data types. While many of these approaches have been in existence, the penetration of innovative data sources that are diverse indicates that data should be widely comprehended and utilized by decision-makers.<\/p>\n
The research maintains that attempting to precisely describe data science boundaries is not important. Data science programs are being developed in various academic fields; thus, such discussions can be done within educational boundaries (Drucker, 1967).\u00a0 For data science to serve organizations, it is essential to comprehend how it relates to other essential and closely correlated concepts and begin understanding the major principles that define data science. In the research, a perspective that provides all the answers to the questions is provided. The first step involves the dissemination of the closely correlated principles. This involves showing data science as the link between data-instigated decision-making and processing technologies like big data. There is also the discussion of complex attributes of data science as a profession and data science as a discipline.<\/p>\n
Significance of the Study<\/p>\n
The study is important since it provides conceptual needs for data-driven organization roles since the variable tasks indicate the processes done when creating value (Leavitt 1965). Decision-making and data tasks can be categorized into horizontal and vertical procedures. Vertical procedures describe the allocation of financial resources and talent, and horizontal procedures allow decision-making and tasks within the functional borders (Sandrin et al., 2014). A successful data-driven organization has to deal with challenges that touch on data shortage, infrastructure, investments, and human resources (Victor & Farkas, 2018). Incorporating procedures and roles is vital in mitigating data analytics and big data costs. Adding engagements that touch on data lifecycle into daily routines enhances an organization’s ability to manage data and data velocity (Shamim et al., 2019). Data analytics creates value in businesses through tasks and procedures that change data into a usable form to deal with issues and provide additional knowledge (Anwar et al. 2018). <\/p>\n
Creating and applying data-driven trends, procedures, and tasks develop a data-driven culture that affects the working environment, leadership techniques, behavior, and strategy (Shamim et al., 2019). A look at some of the clear tasks performed by data-driven organizations is data-driven innovations (DDI) and data-driven business models (DDBM). DDI is used to locate business insights that were previously unknown since they do not result in monetization. Data-driven business models (DDBM) are meant to use data to create value. Whether or not both tasks result in value, it is evident that both have an effect on a business’s performance. People, in this research, considered the variable, means the human resources within a business. To sufficiently engage in different tasks, organizations need a suitable integration of human resource strategies to get talent, create skills, and develop the required mindset to apply the selected data management command and organizational data abilities (Sandrin et al., 2014). High-tech companies need an efficient data management workforce at all levels to maintain sufficient competencies for various synergies, communication, and decision-making (Kates & Galbraith 2007). A value that is data-driven creates managerial hurdles that relate to talent acquisition and development, leadership, and innovation management. Data management abilities are correlated to leadership’s perception of change and the meaning of a data-driven entity. <\/p>\n
With data availability increasing globally, data-driven organizations should include data experts since data management calls for various skills in storing, visualizing, cleaning, and sharing (Shamim et al., 2019). Organizations should have data experts to collect and assess previous and current consumer and market data since many organizations fail since they lack the necessary skills to effectively gather, disseminate, and use data management understandings (Anwar et al., 2018). Research done by Dubey et al. (2019) indicates the importance of human resources in businesses since human skills are important when analyzing big data. The significance of human skills in data analytics is seen when those hired for their knowledge with numbers or have learned to understand their significance are equipped with the best quantitative tools and evidence. This results in some of the best decisions (Davenport, 2006). The positive impacts of data modeling and analytics for organizational roles like pricing, supply chain, or financial support are numerous (Govindan, 2018). Effective resource allocation in terms of funding and workforce increases the chances of success. Even though a lot of investment is required, a knowledgeable workforce that is well-equipped and well-educated is an important success factor in such organizations.<\/p>\n
Another variable in this research indicates the technology used in carrying out responsibilities within the company (Saeed & Wang 2013). To change an organization to a DDO, there is a need for technology that provides the infrastructure that allows for processing and automated engagements like data creation, visualization, or storage. Technology includes the software and hardware needed to sustain the procedures and create value in data-driven organizations (Ashrafi et al., 2019). The study also shows the significance of IT alignment and IT investment in a company’s performance, especially in data-driven organizations, since there is a need for analytical techniques and tools to perform and support data management and data science endeavors, like storage and visualization (Dubey et al. 2019). This increases pressure on IT departments to ease access to data. It also affects decision-making since those working can arrive at meetings ready with their assessments (Berndtsson et al., 2018). Based on a technological perspective, data interaction and visualization are more than just the technological manner of displaying data to a business. They are important in the communication process since it allows those without technical abilities in the business world to participate in conversations regarding data (Vidgen et al. 2017). The organizational structure gives a clear picture of the system and its hierarchy (Wang, 2013). Even though businesses can create data management techniques by acquiring new talent or offering training sessions to their current workforce, leadership should acknowledge its significance as a resource that can improve return on investment (Henderson, 2017).<\/p>\n
The organizational framework indicates the position of authority, power, and leadership since they are created around consumers, products, geographies, and functions (Kates and Galbraith 2007). Consequently, the framework creates a specific role that is hierarchical (Grant 1996). Roles that are clearly defined set the decision-making and hierarchy within a company. Some of the many roles that should be defined to perform and implement analytics within a company include project management, data analysts and scientists, data architects, and engineers (Gandomi & Haider, 2015). The creation of data-driven decision-making and data roles creates a data-driven organizational culture that organizations should utilize (Schuritz et al., 2017). To transform an organization into a data-driven entity, various factors should be considered. The company should create a digitization department that creates strategies and assesses the company’s progress using tools like key performance indexes to be data-driven. Also, the department can create investments that enable the hiring of skilled employees and the use of innovation (Jurgen, 1983). Also, to structure a company’s analytics abilities, companies should have analytics teams that do data analysis and convey the information through all the departments involved (Polzonetti & Sagratella 2017).<\/p>\n
This study’s effect on science indicates a modern-day concept that can support the various factors involved in data-instigated decision-making since classical theories cannot support these factors (Pomerol &\u00a0Adam, 2004). The concept can act as a basis for novice studies and developments in various fields that touch on machine and human collaboration, metahuman systems, and business data analytics (Paschen et al., 2020). Additionally, the research help societies and businesses by showing the elements needed for data-instigated decision-making and how they can result in quality decisions that are informed. Decision-makers and managers can find this valuable when applying machine learning and data analytics in decision-making. This is a simulated analysis of what can be achieved by incorporating the different factors of modern data-driven decision-making (Mandinach, 2012).\u00a0However, this demands an immense amount of effort to get a substantial integration between these factors. The level of collaboration between a machine and human, the appropriate selection of relevant data, and the proper application of analytics techniques and tools, and the assessment of data-instigated decisions and their integration are important aspects that should be studied. Also, the circumstances that result in the rejection or acceptance of data-instigated decisions should be studied (Burton et al., 2020).<\/p>\n
Moreover, the significance of topics like traceability, reliability, transparency, and accountability have always been the subject of intense questioning when coming up with decisions, and their relevance in regard to data-instigated decision-making cannot be overlooked. There are several topics that touch on the integration of machines and humans. With the universal application of assisted AI systems that develop human decision-making, the limited trust in their predictions should be addressed. This calls for studies that consider AI decisions transparently by giving explanations that help in developing trust (Schmidt et al., 2020). Also, with the rise in metahuman systems, there are changes in studies focusing on machine learning and human systems and the key variations exhibited (Lyytinen et al., 2017). Also, future studies should assess the principle of collaborative rationality since incoming decision-making processes will rely on this collaboration (Simon, 1997).\u00a0 <\/p>\n
The study initiatives answer various aspects expected to create the next phase of technology-enabled organizations that create value (Bughin et al., 2010). For example, to maintain distributed innovation on the computer network, there should be growth in distributed storage, workflow sharing, interdisciplinary collaboration, and mobile computing to create resources and spatiotemporally creators. Assessments are increasingly reliant on cloud computing and big data simulations to handle difficulties in the engineering perspectives of complicated systems like logistics, healthcare, and manufacturing (Xu et al., 2015). This needs a knowledgeable workforce, integration across various domains, data processing, and data mining analytics. Acquiring resources to encourage improvement calls for support from various studies since factors like spatiotemporal collocation are important in accomplishing the various innovations using tools, methodologies, and solutions. An example is multi-scale collaborations call for various spatiotemporal levels of integration in various domains maintained by distributed storage. <\/p>\n
The 5 Vs that define big data and the factors of could computing frequently play a key role in the innovation procedures of digital science: the existing innovation possibilities and study agenda of applying cloud computing for handling big data. Various innovative features are coming up as big data expands and becomes popular, especially in relation to how people live, think, and achieve. There is the expectation that the study directions and spatiotemporal thinking described in the study will guide incoming innovation and new businesses in various domains.<\/p>\n
Research Questions<\/p>\n
Some of the research questions are what the key factors of a data-driven organization? What are the conceptual needs for creating a data-driven organization? The first question concentrates on the requirements that should be looked at based on science. When looking at the findings from the first research question, the second research question looks to identify the strategic requirements of a data-driven organization. To find answers to the research questions, a review is done to locate the appropriate literature concerning the phrase data-driven organization. Based on the found reviews, a joint structure is created to indicate key factors of a data-driven organization by using Leavitt’s model to get the conceptual needs from the variables to find answers to the second research question. Even though explanations and solutions have been investigated for various principles in data-instigated decision-making, a comprehensive theory that relies on the scientifically proven concepts of decision theory that integrates and describes the correlation between the factors has not been proposed. Thus the research seeks to address the existing lack of theory in providing contemporary elements of data-instigated decision-making. Thus another research question is how entities can include decision theory to sustain data-instigated decision-making with analytics and data. <\/p>\n
Big data is gathered in a geographical and globally distributed manner. A fast network that links storage to effectively handle big data requires an effective distributed storage method (Aydin et al., 2015). The growth of various processing solutions and distributed storage has developed people’s comprehension and increased the emphasis on acquiring improved support systems in digital science with spatiotemporal traits (Yang et al., 2015). There are various research questions that are considered in improving distributed storage systems. Do the questions include how to link and apply distributed storages to accomplish serialization in areas that are characterized by geographically scattered storage systems? The manner in which an entity can optimize various emerging and traditional database model systems in distributed storage? How to improve performance and data backup in cloud systems and the security alternatives in a cloud environment? How to distinguish and designate big data into various storage solutions like caches and hard drives that are important for enhancing system performance? How can mobile storage be augmented with cloud storage using methods that ensure improved management and the application of distributed storage in cloud and mobile devices? The last research questions look at how to develop smart storage devices that conduct or process data segments on the storage applying co-locations?<\/p>\n
Theoretical Framework<\/p>\n
One of the initial purposes of strategic management studies is to comprehend the unpredictable impacts of strategy on a company’s performance (Mintzberg, 1975). The contingency theory maintains that no singular effective strategy exists for all companies (Zott & Amit 2008). thus the main purpose of a contingency method in organizational theory is to shift its framework to internal capabilities and external sources. Thus, a contingency model in an organization’s ability recognizes decision factors that relate to organizational design (Victer, 2020). Organizational design is a very appropriate analytical method for comprehending a company’s performance (Levitt et al. 1999). The processes involved in decision-making for integrating and differentiating parts that make up the primary influences in strategy implementation impact a company’s performance and structure as various approaches to developing and structuring businesses result in various outcomes under different situations (Tushman & Nadler 1978). Nahapiet and Ghoshal (1998) state that organizations are basically developed around the notion of specialization, differentiation, interdependence, and integrations. This makes the research apply organizational design theory to the study. There have been various structures and methods to organizational design (Niederman & March 2019).<\/p>\n
An example is a six-variable model created by Weisboard (1976) to classify organizational tasks as formal or informal. For an organization to succeed based on this model, it has to be effective in its operations (Saeed & Wang 2013). The congruence model is considered effective in aligning the objectives and requirements of particular organizational factors (Tushman & Nadler 1978). A five-variable model, proposed by Galbriath (2016), can also be used in organizational strategy and implementation. The first considers the strategy which influences a company’s direction, a second is structure that defines where the decision-making power is located, third is the procedures that indicate how information flows, the fourth touches on results that impact people’s motivation within an organization, and the fifth is human resource which describes the employees’ skills and mind-set. <\/p>\n
In its strategy to assess organizational data management and design, the Leavitt model seems appropriate. The model provides a four-variable tool to assess different functions and the correlations between the variables, tasks, structure, people, and technology. In the design, the structure means the location and system of power, the company’s communication methods, and the organization of tasks (Mintzberg, 1989). The variable task means all functions and tasks that offer goods and services to create value. The technology variable includes the applications and hardware used in different tasks. The variable people means human resources in a company that does the tasks (Saeed & Wang 2013). The model is shaped to emphasize the correlation between the variables since a shift in one impacts the organization and other variables (Falletta 2014). The research applies the model to assess the data management needs companies should possess to accomplish various needs and challenges to transform into data-driven organizations and improve their value creation using data (Olszak & Zurada, 2019).<\/p>\n
With key developments in innovation, the rise of AI, analytics, and big data, and the focus on automated and data-driven procedures, indicates the need for novice theory in decision-making based on these situations arises (European Commission, 2020). Though the nature of data-instigated decision-making is, like many decision-related subjects, based on decision theory, it has outgrown the abilities of the past and requires new methods to sustain future scientific studies. Thus, after reading various kinds of literature, a contemporary data-driven theory should be developed. The theory should seek to offer support to decision-makers in the digital fields to get more optimal, data-driven, informed choices. The need to develop the concept in research will be explained, and the reason for practice. This is showcased by examining the decision theory’s opinions in supporting data-instigated decision-making by assessing their disadvantages in discussing data-driven failures and successes.<\/p>\n
Limitations of the Study Assumptions<\/p>\n
Since the research structures and assesses the appropriate study field and this results in many research impacts, it is essential to look at some of the study’s limitations. To begin, it suggests a platform that creates an understanding of what a DDO is and the major factors that result in its creation. Each factor indicates an opportunity for deeper study projects that are valid, correlated, and contextualized within the structure. Though the structure requires high-level work, it allows one to look at DDOs with fewer complications. That is a limitation since there are details that are obscured, thus limiting one’s contribution. <\/p>\n
Other limitations within the study are found in the general aspect of the findings produced. Despite the argument on the value they create, they provide a framework into the topic both for studies and implementation in the work environment. Though the research sought to locate a complete sample, some articles may not have been found in the structured literature assessment. Also, the results, at this point, are fully dependent on data found in various kinds of literature. This stretches the methodological aspects; for instance, by combining experimental assessment into DDOs using questionnaires, interviews, or case studies in organizations that have transformed into data-driven entities, there are limitations and commitments for further studies. By gathering field data from companies undergoing transformation, the key factors could be applied in defining the movement between important elements that should be categorized as consequences or antecedents.<\/p>\n
The decision theory has its limitations when it comes to influencing decision-making tasks and is substandard in adequately showing the reasoning on decisions that touch on various knowledge and communication on the reasons for the selections in ways that all human decision-makers can comprehend. This impedes its application in automated processes since it would call for qualitative objectives, techniques, and preferences and base these selections on analytic and theoretic data instead of ad hoc guidelines (Doyle & Thomason, 1999).<\/p>\n
There are six factors that are used to evaluate the contentious framework of theories, recommendations, and suggestions. The first three are important in any discussion. They include the claim that indicates the argument’s purpose or facts, the warrant that directly or indirectly supports the basis and correlations to the claim. The remaining three factors can be deemed necessary but not important (Karbach, 1987). Thus, the backing is the factor relied on to substantiate and create relevance and reliability of the warrant. Qualifiers are those that impede the argument’s strength or suggest conditions that make the warrant valid, and rebuttal supports situations and exceptions that may make the supporting arguments or claim invalid (Wale-Kolade et al., 2013).<\/p>\n
Additionally, when combining machine learning and human factors in decision-making, many entities avoid failure by maintaining the human factor since it is difficult to depend solely on the machine. The machine’s decision-making power is restricted when looking at the claim and only works as support instead of a replacement for the human factor. Also, the decision-making procedures are clearly stated and adhered to even by the machines when there are fewer alterations. Even though the case may fail to indicate the full extent that can be achieved by the correct implementation of such decision-making techniques, the theory that various factors should exist and a balance created between them is validated. <\/p>\n
Since experiments rarely take place in studying these initiatives, they need random assignments, which means that there is no single strategy that all organizations can use since they are unique. Many also feel that it is tiresome to participate in this type of research though there is always a constant increase of research materials concerning data science, big data, and organizations. Future studies should look at the following questions; does the integration rely on the problem type? Is the correlation between the decision and the human limited to specific decisions like strategy, duration, and techniques? How can the choices be justified since algorithms are famous for their black-box aspects, and how can they be solved? Who or what takes the blame for the wrong decisions made in this scenario? Mistakes in decision-making were created distinctively by machines and humans; now, what type of new mistakes and difficulties developed with their integration? How can humans be taught in analytics and data with a focus on important statistical ideas like errors, accuracy, and uncertainty? Finally, how to create data-driven decision-making that ensures humans remain in the loop and enable them to semi-assess the decision-making procedures instead of being controlled by the algorithms? Thus all these implications should be widely explored before this type of decision-making can be fully optimized. <\/p>\n
The velocity and volume hurdles of big data call for the creation of virtual machines. Independent discovery of the velocity for the provisions of virtual machines is important and should take into account the high efficiency in executing tasks and the optimal expenses involved (Pumma et al., 2012). Studies are being done to comprehend the use and relevance of big data shifting trends to create a complete model that forecasts system behavior as the application trends and working loads undergo constant changes (Castiglione et al., 2014). An instance is a proposal that automated methods should be used to designate the optimal amount of resources in cloud computing. In contrast, other proposals call for an adaptive structure that allows for up and down scaling and the effective utilization of computing resources. <\/p>\n
Also, most task predictions and models used in resource-allocation and algorithms offer or enhance auto-provisioning abilities and auto-scaling. For example, applying the two streaming techniques to assess workload forecasting approaches to analyze the workload attributes can result in significant improvement. The nodes-scheduling model that relies on a chain prediction for assessing big data asserted the deadline as the hindering constraint and suggested a way of assessing the number of resources required to catalog various tasks while considering the data transfer expenses and implementation. Further studies should focus on improving auto-provisioning and auto-scaling abilities in cloud architecture to deal with challenges that affect big data. For instance, improved virtual machine provision techniques should be developed to create a concurrent implementation of many applications.<\/p>\n
Definitions<\/p>\n
In this section, the research will define some of the keywords used to better understand the terms. <\/p>\n
Data-driven Organization, <\/p>\n
A data-driven organization is an organization that understands the significance of gathering unprocessed data and comprehends that choices cannot be made based solely on raw data. Instead, being data-driven entails a deeper assessment of the data collected, cleaning it, and using the information derived from the data to spur development and profitability. It requires organizations to work with the relevant type of data when required. This could be achieved in various ways like assessing client behavior, assessing demographic data, or collecting survey responses. <\/p>\n
Structured Literature Review<\/p>\n
A structured literature review recognizes, chooses, and critically analyzes the study to answer an expressed issue (Dewey & Drahota, 2016). The structured review should adhere to a clearly outlined criteria or strategy in which the protocol is stated before conducting the review. It is a transparent and complete quest done over many datasets and literature that other researchers can duplicate and duplicate. It entails designing an effective plan that has a particular or provides answers to a specified question. The review distinguishes the nature of the information examined, critiqued, and summarized with known durations. The search phrases, strategies, and limits have to be incorporated in the review. <\/p>\n
Some of the key concepts behind a structured literature review include transparency, focus, integration, clarity, equality, coverage, and accessibility (Pittway, 2008). <\/p>\n
Data Management<\/p>\n
Data management involves ingesting, saving, creating, and sustaining the data collected or created by a company. Effective data management is important in expanding the information technology systems that drive applications in businesses and offer analytical knowledge to help run operational decision-making and planning by leaders and end-users. The data management procedures entail a blend of various roles that collectively seek to ensure that the data in an organization’s system is precise, convenient, and accessible. Most of the work needed is carried out by data management and IT teams, even though users can also engage in various parts of the procedure to guarantee that the data is according to their needs and they adhere to the policies dictating its utilization. <\/p>\n
Data Analytics<\/p>\n
Data analytics is described as the science of assessing unprocessed data to come up with conclusions on the information. Various data analytics methods and procedures have been transformed into mechanical procedures and algorithms that work on raw data for human use. Data analytics helps organizations improve performance. The term is broad since it consists of various types of data assessment methods to understand how things can be improved. The techniques can show metrics and trends that would be left in mass information. The information is applied in the optimization of procedures to improve a business’s efficiency (Strand &\u00a0Syberfeldt, 2020). It is essential since it assists businesses in enhancing their performances. Applying it to a business model indicates organizations can assist in mitigating costs by distinguishing more effective ways of engaging in business and by keeping vast amounts of data (Guggenberger et al., 2020). An organization can apply data analytics to improve its decision-making processes and assess consumer trends and levels of satisfaction, thus resulting in new and improved products or services. <\/p>\n
Data-driven decision-making.<\/p>\n
Data-driven decision-making is described as the use of metrics, facts, and data to develop strategic decisions relevant to a company’s objectives and initiatives. When companies understand the complete extent of their data, it means that various entities within the organization are empowered to make better decisions daily. This is not accomplished by only selecting the right analytics innovation to recognize strategic opportunities. A company should ensure data-driven decision-making is used every day to develop a culture that fosters curiosity and critical assessments. Those at different levels within an organization should ensure their conversations begin with data, and they should engage in constant practice and utilization to improve their data techniques. Foundationally, this calls for a model that allows people to get a hold of the data they feel is relevant with the utmost guidance and security initiatives. There is also the need for development opportunities that allow the workforce to comprehend data techniques. Lastly, possessing a community and advocacy that makes and sustains data-driven decisions will make others use these techniques (Habermas, 1984).\u00a0 <\/p>\n
Big Data Analytics<\/p>\n
Big Data analytics means using excellent analytic methods in assessing vast amounts of data sets that entail structured and unstructured data derived from various sources and in varying sizes (Elgendy &\u00a0Elragal, 2014).\u00a0 This type of analytics is a procedure used to derive important insights like hidden trends, unknown links, consumer preferences, and market patterns. It can be used for improved decision-making, mitigating fraudulent deeds, just to name a few. Big data analytics defines the procedures of revealing patterns and links in vast quantities to assist in making informed decisions (Saggi &\u00a0Jain, 2018). The procedures utilize simple statistical assessment methods like regression and clustering and use them in large datasets using innovative tools and machines. Big data has been a common term since hardware and software abilities made it possible for companies to apply vast amounts of raw data. The growth of new technologies has led to more data becoming available to companies. The field continues to experience constant growth as data entities search for methods of incorporating the huge amount of complicated data created by transactions, sensors, and networks. Big data analytics are being used with innovations like machine learning. <\/p>\n
Automated Decisions<\/p>\n
Automated decisions are procedures of coming up with decisions without the involvement of a human being. The decisions can be based on real data, digitally developed profiles, or assumed data. Some of the examples of automated decisions include a website’s decision to provide a loan and aptitude assessments applied in recruitment processes that apply algorithms. These types of decisions usually entail profiling, but it is not a must. <\/p>\n
Decision Theory <\/p>\n
The decision theory is the assessment of an agent or person’s decisions. The concept helps in comprehending the reasoning behind the decisions various people make. The selections usually have impacts and are usually described in unique branches. When assessing decision theory, assessments always contain the things that result in optimal decisions, the attributes of the decision-makers, and the manner in which the decisions were made. Describing how entities should make decisions in specific situations is also part of the assessment. The concept of decision theory is usually discussed in fields like statistics and economics, and in this research, it is used in analytics (Kahneman, 2003).\u00a0<\/p>\n
Algorithmic Decisions<\/p>\n
Algorithmic decisions utilize statistical and data analyses to categorize individuals or the reasons for analyzing their qualifications to either gain or gt a penalty. These types of systems are used in credit decisions, and in recent years they have been applied in insurance qualification, marketing, and employment screening. They are also used in the public domain, including delivering services, and in the justice corridors, it helps in sentencing. Many of the automated decisions depend on statistical methods like regression assessments. Recently, though, the systems have included machine learning to enhance their fairness and accuracy. These statistical methods try to locate trends in data without the need for analysts to define the factors they can use. They will often get new random and not obvious connections to the analysts or apply a theoretical comprehension of the subject matter. Because of this, they can assist in discovering novice elements that enhance the precision of predictions and selections. <\/p>\n
Data-Driven Business Models.<\/p>\n
Organizations with data-driven business models rely on data for their main operations. The dependence or focus on data can impact various dimensions like the added value and the value proposition. Value added is created from data by making it a major resource in organizations. This indicates that its main pursuits are data utilization, acquisition, and assessment. Innovation is offering more organizations and industries increased opportunities to apply the vast amounts of data created and apply it in their operations, for instance, to execute smart products using wireless sensors or combine people into the decision-making procedures using mobile computing. All solutions that are data-driven indicate that when data is incorporated in the key elements of an organization’s model, it goes far from assessing and applying data to make the value creation process more effective. Organizations should ask themselves how they can apply technology and existing data to create and deliver services and products that are data-driven.<\/p>\n
Summary<\/p>\n
To conclude the chapter, previously, studies began detailing decision systems, and this included the decisions people make alongside machines and software that had enhanced predictive abilities. This increased the interest in decision systems in terms of systems, people, processes, and data for sustaining decision procedures and making autonomous decisions. Also, with improvements in machine learning, data science, analytics, data-instigated decision-making based on findings and evidence indicated through analytics has increased in popularity. This has made data-driven decision-making be considered as a way of providing quality decisions based on facts since it integrates human experience with data analytics to create room for rational decisions, thus improving outcomes (Kalantari, 2010).<\/p>\n
Big data has developed over the past few years as a new concept that offers numerous opportunities and data to enhance studies and decision-support mechanisms with unparalleled value for innovations in sciences, businesses, and engineering. Concurrently, big data has its complications when it comes to storage, movement, processes, extraction, and the purpose of data. Breakthroughs like cloud computing offer significant support to deal with the challenges of common computing resources like networking, computing, and storage software (Storm & Borgman, 2020). Such breakthroughs have led to big data developments. The paper will also look at two perspectives, cloud computing, and big data, and assess the gains and risks of applying cloud computing in handling big data in various science disciplines. Based on the introduction, hurdles, sources, research opportunities, and innovation status, the following conclusions are made. Big data and cloud computing encourage application developments and scientific discoveries. Cloud computing is an important factor in solving the storage problem associated with big data. <\/p>\n
Even though the concept of cloud computing was formed before big data, it is considered an innovative paradigm for achieving computation with attributes such as pooled resources, self-service, elasticity, on-demand access, and pay-as-you-go. The traits allow cloud services to be considered as infrastructure, platform, and software services (Hassan &\u00a0Mingers, 2018).\u00a0 Having a role in defining the opportunities of different disciplines, cloud computing and big data have resulted in improved solutions for various problems in various disciplines like business, social science, industry, and astronomy. <\/p>\n
Also, the increasing need for precision, big data, and improved resolutions have encouraged the development of cloud computing and related innovation. Cloud computing offers innovative methods of doing business and supporting creativity. Bringing together the economy, big data, and computing of services and goods have encouraged various conversations on IT-related concepts since they are part of everyday purchasing procedures. There are proposals that the big data entities that have 5V features and issues will continue encouraging the rapid development of associated cloud computing innovation in various perspectives.<\/p>\n
The study attempts to determine significant factors of data-driven businesses and their conceptual needs. As indicated by Webster and Watson (2002), a literature assessment was conducted. To answer the research questions, the study uses a joint structure showcasing significant factors of data-driven organizations and their attributes. Conceptual needs derived from various studies were constructed and based on organizational design theory to determine the research questions. The research assesses and structures related study concepts and is followed by many implications. <\/p>\n
Of significance is the fact that the research can act as a checklist for important aspects of the transformation procedures. Uniquely, the structure addresses managerial positions that concentrate on data, like the Chief Data Officer position to guide the transformation procedures effectively. Also, the research provides a descriptive value that indicates the attributes of a data-driven organization and provides suggestions on the areas that should be considered with the utmost care. <\/p>\n
Also, the research intends to answer the identified research questions by looking at various kinds of literature to create a decision theory that accounts for the abilities of data-instigated decision-making by incorporating the decision-making factors with current advances in data analytics and big data. The theory seeks to act as an epistemological foundation that supports the engagements of data-instigated decision-making by offering explanations beyond the theory, thus allowing for future studies in this field. To circumvent common gap-spotting assessments and carry out innovative research, the concept is explained using a study known as problematization as a methodology by Alvesson and Sandberg (2013). This calls for the application of various methodological concepts, including recognizing a field of literature, recognizing and explaining assumptions that characterize the field, assessing the assumptions, creating an alternative, looking at how it relates to the audience, and assessing the alternative assumptions. <\/p>\n
The research maintains that attempting to precisely describe data science boundaries is not important. Data science programs are being developed in various academic fields; thus, such discussions can be done within educational boundaries. For data science to serve organizations, it is essential to comprehend how it relates to other essential and closely correlated concepts and begin understanding the major principles that define data science. In the research, a perspective that provides all the answers to the questions is provided. The first step involves the dissemination of the closely correlated principles. This involves showing data science as the link between data-instigated decision-making and processing technologies like big data. There is also the discussion of complex attributes of data science as a profession and data science as a discipline.<\/p>\n
The research is vital since it provides conceptual needs for data-driven organization roles since the variable tasks indicate the processes done when creating value. Decision-making and data tasks can be categorized into horizontal and vertical procedures. Vertical procedures describe the allocation of financial resources and talent, and horizontal procedures allow decision-making and tasks within the functional borders. A successful data-driven organization has to deal with challenges that touch on data shortage, infrastructure, investments, and human resources. Incorporating procedures and roles is vital in mitigating data analytics and big data costs. <\/p>\n
Creating and applying data-driven trends, procedures, and tasks develop a data-driven culture that affects the working environment, leadership techniques, behavior, and strategy. A look at some of the clear tasks performed by data-driven organizations is data-driven innovations (DDI) and data-driven business models (DDBM). DDI is used to locate business insights that were previously unknown since they do not result in monetization. Data-driven business models (DDBM) are meant to use data to create value. Whether or not both tasks result in value, it is evident that both have an effect on a business’s performance. People, in this research, considered the variable, means the human resources within a business. To sufficiently engage in different tasks, organizations need a suitable integration of human resource strategies to get talent, create skills, and develop the required mindset to apply the selected data management command and organizational data abilities. <\/p>\n
Another variable in this research indicates the technology used in carrying out responsibilities within the company. To change an organization to a DDO, there is a need for technology that provides the infrastructure that allows for processing and automated engagements like data creation, visualization, or storage (Sch\u00fcritz et al., 2017). Technology includes the software and hardware needed to sustain the procedures and create value in data-driven organizations. The study also shows the significance of IT alignment and IT investment in a company’s performance, especially in data-driven organizations, since there is a need for analytical techniques and tools to perform and support data management and data science endeavors, like storage and visualization. This increases pressure on IT departments to ease access to data. It also affects decision-making since those working can arrive at meetings ready with their assessments.<\/p>\n
Do the research questions include what the key factors of a data-driven organization are? What are the conceptual needs for creating a data-driven organization? The first question concentrates on the requirements that should be looked at based on science. When looking at the findings from the first research question, the second research question looks to identify the strategic requirements of a data-driven organization. To find answers to the research questions, a review is done to locate the appropriate literature concerning the phrase data-driven organization. Based on the found reviews, a joint structure is created to indicate key factors of a data-driven organization by using Leavitt’s model to get the conceptual needs from the variables to find answers to the second research question. Even though explanations and solutions have been investigated for various principles in data-instigated decision-making, a comprehensive theory that relies on the scientifically proven concepts of decision theory that integrates and describes the correlation between the factors has not been proposed. Thus the research seeks to address the existing lack of theory in providing contemporary elements of data-instigated decision-making. Thus another research question is how entities can include decision theory to sustain data-instigated decision-making with analytics and data. <\/p>\n
Big data is gathered in a geographical and globally distributed manner. A fast network that links storage to effectively handle big data requires an effective distributed storage method. The growth of various processing solutions and distributed storage has developed people’s comprehension and increased the emphasis on acquiring improved support systems in digital science with spatiotemporal traits (Jha &\u00a0Ngai, 2020). There are various research questions that are considered in improving distributed storage systems. Do the questions include how to link and apply distributed storages to accomplish serialization in areas that are characterized by geographically scattered storage systems? The manner in which an entity can optimize various emerging and traditional database model systems in distributed storage? How to improve performance and data backup in cloud systems and the security alternatives in a cloud environment? How to distinguish and designate big data into various storage solutions like caches and hard drives that are important for enhancing system performance? How can mobile storage be augmented with cloud storage using methods that ensure improved management and the application of distributed storage in cloud and mobile devices? The last research questions look at how to develop smart storage devices that conduct or process data segments on the storage applying co-locations?<\/p>\n
In its strategy to assess organizational data management and design, the Leavitt model seems appropriate. The model provides a four-variable tool to assess different functions and the correlations between the variables, tasks, structure, people, and technology. In the design, the structure means the location and system of power, the company’s communication methods, and the organization of tasks. The variable task means all functions and tasks that offer goods and services to create value. The technology variable includes the applications and hardware used in different tasks. The variable people means human resources in a company that do the tasks. The model is shaped to emphasize the correlation between the variables since a shift in one impacts the organization and other variables. The research applies the model to assess the data management needs companies should possess to accomplish various needs and challenges to transform into data-driven organizations and improve their value creation using data. <\/p>\n
With key developments in innovation, the rise of AI, analytics, and big data, and the focus on automated and data-driven procedures, indicates the need for novice theory in decision-making based on these situations arises. Though the nature of data-instigated decision-making is, like many decision-related subjects, based on decision theory, it has outgrown the abilities of the past and requires new methods to sustain future scientific studies. Thus, after reading various kinds of literature, a contemporary data-driven theory should be developed. The theory should seek to offer support to decision-makers in the digital fields to get more optimal, data-driven, informed choices. The need to develop the concept in research will be explained, and the reason for practice. This is showcased by examining the decision theory’s opinions in supporting data-instigated decision-making by assessing their disadvantages in discussing data-driven failures and successes.<\/p>\n
Incoming studies should look at the following questions; does the integration rely on the problem type? Is the correlation between the decision and the human limited to specific decisions like strategy, duration, and techniques? How can the choices be justified since algorithms are famous for their black-box aspects, and how can they be solved? Who or what takes the blame for the wrong decisions made in this scenario? Mistakes in decision-making were created distinctively by machines and humans; now, what type of new mistakes and difficulties developed with their integration? How can humans be taught in analytics and data with a focus on important statistical ideas like errors, accuracy, and uncertainty? Finally, how to create data-driven decision-making that ensures humans remain in the loop and enable them to semi-assess the decision-making procedures instead of being controlled by the algorithms? Thus all these implications should be widely explored before this type of decision-making can be fully optimized. <\/p>\n
The velocity and volume hurdles of big data call for the creation of virtual machines. Independent discovery of the velocity for the provisions of virtual machines is important and should take into account the high efficiency in executing tasks and the optimal expenses involved. Studies are being done to comprehend the use and relevance of big data shifting trends to create a complete model that forecasts system behavior as the application trends and working loads undergo constant changes. An instance is a proposal that automated methods should be used to designate the optimal amount of resources in cloud computing. In contrast, other proposals call for an adaptive structure that allows for up and down scaling and the effective utilization of computing resources. <\/p>\n
Also, most task predictions and models used in resource-allocation and algorithms offer or enhance auto-provisioning abilities and auto-scaling. For example, applying the two streaming techniques to assess workload forecasting approaches to analyze the workload attributes can result in significant improvement. The nodes-scheduling model that relies on a chain prediction for assessing big data asserted the deadline as the hindering constraint and suggested a way of assessing the number of resources required to catalog various tasks while considering the data transfer expenses and implementation. <\/p>\n
Further studies should focus on improving auto-provisioning and auto-scaling abilities in cloud architecture to deal with challenges that affect big data. For instance, improved virtual machine provision techniques should be developed to create a concurrent implementation of many applications.<\/p>\n
Chapter Two<\/p>\n
Review Of Literature<\/p>\n
Introduction<\/p>\n
Data has become one of the most important assets in most companies. Companies use analytical tools and techniques to get useful insights and information from big data. The companies use this insight and knowledge to understand their processes better and gain a competitive advantage (Kamioka &\u00a0Tapanainen, 2014). Big data was always associated with large companies; however, even small companies do have big data. The difference between a successful company and an unsuccessful company is that successful companies are able to utilize their big data well and use the insights to improve their processes. Even if big data and data science have brought great benefits to companies, there are many challenges associated with big data analytics and data science (Elgendy &\u00a0Elragal, 2016).\u00a0 <\/p>\n
Data is continuously produced and at a very quick pace. It has become an essential asset for every company. The amount and range of data greatly surpassed the ability to do manual analysis and, in some situations, exceeded the efficiency of traditional databases. Machines have become much more efficient at the same time, networking is pervasive, and algorithms have been created that can link databases to allow for wider and deeper research than was possible earlier. Principles of data science and big data analytics are widely used in companies to increase their efficiency. The more database a company has, the more successful that company can be. <\/p>\n
A data-driven company is not about the amount of data collected but about how it effectively uses the collected data. To be data-driven, a company has to gather data and use it as guidance for future decision-making (From Analytics to Action Part I, 2012). A data-driven decision making is decisions made based on analyses of data. A company might follow a different direction than it would have desired when its decisions are made data-driven. Gathering more data is a closer step towards getting to know customers’ mind and their behavior. <\/p>\n
Big data analytics deals with the analyses of big datasets. Big Data Analytics represents intelligence problems that are too large, too disorganized, and too quick to be solved by conventional approaches. Big Data provides substantial new possibilities for companies to extract new value from their most important asset: data and to generate market advantage. The overall valuation of big data was estimated at $6.3 billion as of 2012, according to a new industry study released by Transparency Market Research, but by 2018 it is projected to hit the incredible amount of $48.3 billion, which is almost a 700 percent increase (Zakir, 2015).<\/p>\n
Literature Review<\/p>\n
The literature review section will tackle specific areas of data science and big data. First, the definition of big data shall be explored in association with data science. This makes it possible to assess the potential of the disciplines regarding growth, availability, and use of information to make apps that suit the needs of the users. An analysis of what data means to organizations, a study of the potential of data in making sense of the analytics shall be analyzed. This means an assessment of insights into how data science helps check its role in organizational routines, processes, and even decisions. <\/p>\n
There is no single accepted definition since Big Data and Analytics is a relatively recent and emerging phrase; different stakeholders have various and often conflicting meanings. In addition to being broad and complicated, Big Data requires advanced technologies to interpret and process. In 2012, McKinsey & Company conducted a survey of 1,469 managers across diverse countries, sectors, and business sizes, in which 49 percent of respondents said their businesses concentrate big data activities on consumer insights, segmentation, and targeting to boost overall efficiency (Zakir, 2015). Wide-scale e-commerce is especially data-intensive since large numbers of clients and transactions are involved. We will quickly present some implementations of the Big Data problems in trade and industry, social administration, and scientific research fields in the following subsections (Tkacz & Kapczynski, 2009). Sixty percent of respondents said their firms should concentrate on using data and analytics to produce these insights. Only one-fifth said their companies have completely implemented data and analytics to produce insights into one business unit or service. Only 13 percent use data to generate insights around the business. The concern is no longer whether big data can benefit business, as these survey findings indicate, but how a business can achieve full results from big data.<\/p>\n
What is data-driven decision making<\/p>\n
Data-driven decision-making is defined as using facts, indicators, and information to guide business development decisions consistent with our priorities, objectives, and strategies. Data science is the study of concepts, methods, and strategies for understanding phenomena by analyzing data. Data science’s primary aim is to improve decision-making, which is of utmost importance to the industry in general. Data-driven decision-making applies not solely to the perception of data but to the process of forming decisions after analyzing data. Every company aims to be a data-driven industry (Mezias &\u00a0Starbuck, 2009). No company will ever say, “Let’s not use data; we’ll make decisions based on our intuitions.” Data is much essential for the performance of every company. Without using data, companies tend to make incorrect assumptions and biased decisions, which leads to the company’s poor performance. <\/p>\n
In enterprise-level companies and organizations, the processing and review of data have long played an important role. But as society produces more than 2.5 quintillion bytes of data every day, gathering, processing, and interpreting data into actual, actionable information has never been simpler for organizations of all sizes. Although data-driven decision-making has occurred in the industry for decades in one way or another, it is a genuine innovation. In almost every industry, including investment banking, software industries, communications, education, travel and leisure, engineering, retail and social media, data-driven decision making is now prominent (Langley et al., 1995).\u00a0<\/p>\n
Benefits of Data-driven decision making<\/p>\n
Whatever decision is made in a data-driven decision-making process will be more effective as it involves a proper analysis of data and decisions based on the analyzed results.<\/p>\n
Evidence-based decisions often allow historical data to determine what is going to happen in the future. There are several dangers without evidence, such as performing on false beliefs and being influenced by prejudices. The methodology can be used by large companies for diagnostic modeling and processing of big data analysis to boost excellent efficiency.<\/p>\n
Another value of the management of data-based decisions is that it contributes to continuous change. Most companies can incorporate gradual improvements, monitor critical indicators, and make more changes based on data-based <\/p>\n
decision making performance. This boosts a company organization’s average efficiency and efficacy.<\/p>\n
The value of running a data-driven enterprise is that you have continuity over time. The technique requires individuals inside the business to know how choices are made. Individuals should understand the effects of information gathered, evaluated, and controlled, and they take steps accordingly.<\/p>\n
Another value of the data-driven approach to decision-making is that it helps to gain feedback. This helps to study what is meant to be used and what is not. It allows the business to produce innovative materials, production facilities, and new workplace programs.<\/p>\n
Data science and its relationship to big data and data-driven decision making<\/p>\n
According to Provost et al. (2015), with the vast amount of data readily available, companies in every industry are focusing on exploiting and utilizing data for enhancing competitive advantage. The variety and volume of data are outstripping the capacity of manual evaluation, and even exceeding the capacity for the usual database. Computers are becoming increasingly powerful, with networking being ubiquitous and increased development of algorithms. Algorithms are connecting datasets to enable deeper and broader analytics. The growing need for the use of big data analytics and data science in companies is immense. <\/p>\n
Organizations are working towards ensuring that they implement both big data analytics as well as data science in their systems. The growth of these two fields has been contributed by the amount of data that organizations are getting. It is reported that the growth in the amount of data has been contributed by the growth of the internet and the growth of other analytical tools. The whole world is entering into an era of big data. Organizations have realized that data has a great potential of impacting their business processes, which would mean an increase in the efficiency of the business processes. Data is the primary source of knowledge and information on which decisions of organizations are based upon. Data provides insights that are based on facts; thus, organizations have total trust in these insights. Currently, many people have made bid data and data science to be buzzwords. <\/p>\n
The emerging role of data scientists on software development teams<\/p>\n
The authors argue that most people use big data analytics and data science to refer to any analytical tasks. When it comes to big data, it has unique characteristics that differentiate it from other data types. The special attributes of big data make it impossible for traditional analytical methods and databases to handle them. Thus, when an organization decides to implement big data technologies and data science, it should have a well-planned strategy that should help them to be successful. Otherwise, they will not get the maximum benefit from the two. The advancement that occurred on the internet brought about many impacts, one of them is an increase in the amount of data collected. Devices that are collected to the internet collect data daily about the user. Some of the data include the date of birth, email address, physical location, age, and gender. <\/p>\n
With the advancement in technology, devices are now connected, which are able to collect, store, and relay data among themselves. Accordingly, the amount of data that companies and businesses collect has tremendously increased over the last couple of years. The data that companies collect are from their customer systems as well as from their machines. Most organizations have realized the potential that data have in increasing their production and leading to an improvement in their overall performance. <\/p>\n
Data-intensive applications, challenges, techniques and technologies<\/p>\n
An algorithm is a set of rules that proceed in logical steps to aid in solving problems. Algorithmic decision-making systems focus on the analysis of large amounts of personal data to bring out information considered useful in making decisions. Two types of automation are used: tools that get rules from historical data to make predictions; and pre-programmed rules authored by humans. This paper will analyze the challenges that algorithmic decision-making will pose to the rule of law. Algorithmic decision-making systems sometimes fall short of the idea brought out by the rule of law. In China, the rule of law is a traditional legal system that requires that all requirements are followed and enforced. The practice of law in China requires that the government and the legal system as a whole provide for the protection of rights and freedoms, provide for ways to settle disputes following the law, and make fair and impartial decisions about legal matters. In China, algorithmic systems are used by the government to give power to computers to make decisions. These computers can make decisions about child support, taxation, social security, and parental leave. Even the health sector in China stands to benefit from algorithmic systems as surgeries are now performed through robotics that helps to improve the efficiency of operations. Road accidents are far being prevented through the use of technology, such as collision avoidance sensors. <\/p>\n
Mobile networks and applications<\/p>\n
The algorithmic decision-making system poses a challenge to the rule of law in the sense that it is difficult to know what is happening as many programming entities are reluctant to disclose the coding program they have used because of commercial reasons and competition in the business. Governments are hesitant to regulate tech companies for fear that they will curtail innovation. In contrast, tech companies are resistant to including human rights into account while designing their programming systems. Access to services is becoming depersonalized as citizens lose their control of the machine systems, thus losing their dignity. The rule of law requires that a citizen’s dignity should be upheld at all costs. Algorithms also tend to put a lot of control in the hands of governments that may lead to biasness and some form of dictatorship. This will be a challenge to the rule of law as it advocates for democracy and freedom of choice for citizens.<\/p>\n
Data-driven dashboards for transparent and accountable decision-making in smart cities<\/p>\n
According to Matheus et al. (2018), data has become one of the most important assets in most companies. Companies use analytical tools and techniques to get useful insights and information from big data. The companies use this insight and knowledge to understand their processes better and gain a competitive advantage. Big data was always associated with large companies; however, even small companies do have big data. The difference between a successful company and an unsuccessful company is that successful companies are able to utilize their big data well and use the insights to improve their processes. Even if big data and data science have brought great benefits to companies, there are many challenges associated with big data analytics and data science. <\/p>\n
While the potential benefits of big data analytics and data science are significant, and initial success has been achieved, there are many challenges that affect the field that must be addressed for the full potential to be realized. One of the challenges of handling large amounts of data is that there are challenges in processing large amounts of data. Large amounts of data need sophisticated technologies that would help to process them. In addition, traditional technologies and database systems are not able to effectively handle large amounts of data. The nature of big data is heterogeneous. This means that big data comes in different formats. Most of the time, the format is semi-structured, unstructured, and structured. This means that the big data has to be converted into a suitable format that would allow smooth data processing.<\/p>\n
Data science life-cycle<\/p>\n
The Data Analytics Lifecycle describes best practices in the analytics process covering exploration to the project’s execution. In the field of data analytics and decision science, the lifecycle draws from existing approaches. After collecting feedback from data scientists and consulting existing methods that provided input on parts of the process, this formulation was generated. Data Analytics Life-Cycle has six phases. It is an overview to test whether it is necessary to stay in the current phase or find whether it is time to move to the next phase.<\/p>\n
Below given is a brief explanation of six phases of the Data Analytics Life-Cycle:<\/p>\n
Phase One: Discovery<\/p>\n
The team studies the business area, including related histories, such as whether similar ventures have been attempted by the company or business unit in the past from which they can benefit. In terms of staff, technology, time, and data, the team analyses the tools required to support the project. At this point, critical tasks include framing the market dilemma as an empirical issue that can be solved in subsequent stages and formulating initial hypotheses to test and begin studying the details.<\/p>\n
Phase Two: Data preparation <\/p>\n
The second phase of the Data Analytics Lifecycle involves preparing data, which requires modeling and analysis to explore, pre-process, and state data. At this point, the team needs to create a robust framework in which the data separated from a production environment can be explored. Typically, this is achieved by planning a sandbox with analytics. The team has to execute ETLT by combining extracting, transforming, and loading data into the sandbox to get the sandbox data.<\/p>\n
Phase Three: Model Planning<\/p>\n
In Phase 3, based on the project purpose, the data science team selects candidate models to add to the clustering data, classifying or finding associations. The team refers to the theories formed in Step 1 at this point when they first got acquainted with the details and understood the market issues or sector. These theories help the team frame the analytics to be done in Phase 4 and choose the best methods to accomplish its aim.<\/p>\n
Phase Four: Model Building<\/p>\n
In Phase 4, for training, research, and development purposes, the data science team needs to create datasets. These datasets help the data scientist to build and train the analytical model while leaving some of the data for model testing aside. It is important to ensure that the training and evaluation datasets are adequately stable for the model and computational techniques during this process. The preparation dataset for performing the initial experiments and the test sets for validating an approach after the initial experiments and models have been performed is a clear way to think about these datasets.<\/p>\n
Phase Five: Communicate Results<\/p>\n
The team has to equate the modeling effects to the parameters set for performance and loss after implementing it. Considering assumptions, conclusions, and any drawbacks of the observations, the team considers how best to articulate the findings and results to the different team members and stakeholders. As the presentation is often shared within an organization, accurately articulating the findings and suitably positioning the audience’s results are important.<\/p>\n
Phase Six: Operationalized<\/p>\n
This is the final phase. The team communicates the advantages of the project more generally in the final phase and establishes a test project to execute the work in a managed manner before extending the work to a full enterprise or customer environment. The team ranked the model in the analytics sandbox in Step 4. Step 6 illustrates the first path to implementing modern analytical techniques or models in a manufacturing context for most analytics teams. Rather than quickly deploying these models on a broad-scale. The risk can be handled more reliably, and the team can learn before a broad scale launch by conducting a small-scale pilot implementation. This technique helps the team learn on a small scale about its success and associated limitations in a manufacturing setting and changes before complete deployment. <\/p>\n
Designing of data-driven apps<\/p>\n
Data-driven models allow designers of mobile applications to consider best practices and patterns and can be used to make design success forecasts and enable the development of adaptive User Interfaces (Deka et al., 2017). Data-driven apps work on various data obtained from many different sites, often in real-time, and generate results entirely different from conventional apps in many different ways. These apps require a lot of testing as it produces an error if any issue with the data is present. These applications are often commonly distributed on various platforms, including handheld devices and normal web browsers, which means they require a framework for distribution that is modular, scalable, and secure. They need to be constantly developed to respond to new use cases or customer demands, considering these applications’ criteria. All changes need to be made online since they have to be accessible every time. Data-driven applications need a large set of data for processing efficiently. A data-driven decision-making model’s primary aspect is to understand the customer’s need through feedback or surveys.<\/p>\n
Data-intensive applications, challenges, approaches, and innovations<\/p>\n
Big Data has altered our way of doing industry, management, and science. Data-intensive research is moving into the world, especially in data-intensive computation, which seeks to provide the resources we need to cope with big data issues. In terms of the preceding three, data-intensive research is emerging as the fourth scientific model, namely, observational science, theoretical science, and computer science. Only because of growing data-intensive technologies has the world of research shifted. The methods and strategies for this form of data-intensive science are entirely different from the three previous ones. Data-intensive research is thus seen as the latest and the fourth model in science for scientific discovery.<\/p>\n
When more and more areas include big data issues, we have reached the Big Data age, ranging from the financial economy to society’s governance and from scientific research to national security. Big Data has a strong connection with e-Science (Hey & Anne, 2002), a computer-intensive science generally applied in distributed computer systems. e-Science, which includes grid computing, can fix several challenges for big data applications (Bart Jacob, Michael Brown, Kentaro Fukui, Nihar. Trivedi Introduction to Grid Computing IBM Redbooks Publication (2005), n.d.). Big data is commonly encountered by web-based apps, such as recent social networking analyses, prediction markets, internet text and records, and internet search indexing. Big Data also offers technology that, such as the Access Grid, facilitate distributed collaboration. Big Data Applications include scientific fields such as physics, atmospheric science, medicine, genomics, genetics, biogeochemistry, and other nuanced and interdisciplinary scientific studies (Philip Chen & Zhang, 2014). <\/p>\n
Big data methods and technologies<\/p>\n
To capture the importance of Big Data, we need to build new analytical approaches and technologies. Until now, a broad range of methods and technology have been developed by scientists to collect, curate, interpret and visualize big data. Even so, they are far from satisfying several requirements (Bell et al., 2009). Several disciplines cover these techniques and technologies, including computer science, economics, mathematics, statistics, and other expertise. To discover useful knowledge from Big Data, multidisciplinary strategies are needed. We will explore emerging approaches and developments (Hey et al., 2009).<\/p>\n
For Big Data to make sense, channels are required. Current resources focus on three types, namely batch processing tools, stream processing tools, and tools for immersive analysis. The Apache Hadoop infrastructure, such as Mahout and Dryad, is based on most batch processing tools. The above is more like the real-time computational prerequisite for stream data applications. For large-scale streaming analytic data platforms, Storm and S4 are good examples. In an immersive world, interactive research processes the data, empowering users to carry out their own data analysis. The consumer is directly connected to the device and can thus communicate in real-time with it. The data may be checked, compared, and evaluated simultaneously in tabular or graphic format or both. Big Data systems focused on immersive analysis are Google’s Dremel and Apache Drill. <\/p>\n
Big Data Techniques<\/p>\n
To effectively process vast quantities of data under short running times, Big Data requires exceptional techniques. Specified applications fairly guide big data strategies. Wal-Mart is an example of using big data techniques, as it employs machine learning and mathematical approaches to analyze correlations from the vast amount of transaction data. In price policies and promotional promotions, these trends may yield greater competition. A variety of fields include Big Data techniques, including analytics, data mining, machine learning, neural networks, the study of social networks, signal processing, pattern recognition, optimization methods, and approaches to visualization (Vassakis et al., 2018).<\/p>\n
Optimization Methods<\/p>\n
Big Data is used to solve complex optimization methods, which require huge memory and time consumption. Optimization methods solve quantitative problems in physics, operation research, operation management, and economics. In several large data applications, including WSNs and ITSs, real-time optimization is also needed. Different methods for optimization problems include data reduction and parallel processing too.<\/p>\n
Statistics<\/p>\n
Statistics deals with the collection, organization, and interpretation of various data. Statistical Techniques are usually used to draw a relationship between various objectives. Standard Statistical Techniques are generally not suited to manage Big Data, and therefore, researchers suggest new techniques or extensions of classical methods. <\/p>\n
Data mining<\/p>\n
Data mining is a series of techniques, including clustering analysis, classification, regression, and correlation rule learning, to extract useful patterns and trends from data. This includes the methodology and statistics of machine learning. In contrast with conventional data mining algorithms, big data mining is more difficult.<\/p>\n
Machine learning<\/p>\n
Machine learning is a significant subjection in artificial intelligence and targets design algorithms that allow computers to evolve behaviors based on empirical evidence. Discovering information and making intelligent choices automatically is the most apparent aspect of machine learning. \u2800<\/p>\n
Visualization Approaches<\/p>\n
The methods used to construct charts, pictures, graphs, and other intuitive view ways to interpret data are visualization approaches. Due to the complexities of 3Vs or 4Vs, Big Data visualization is not so straightforward as conventional relatively small data sets. The proliferation of conventional approaches to visualization is still emerging, but far from enough. When it comes to large-scale data visualization, many researchers use feature extraction and geometric modeling to substantially minimize the data’s size before the final data rendering<\/p>\n
Social Network Analysis<\/p>\n
Social Network Analysis, which in contemporary sociology has arisen as a central methodology, considers social interactions in network theory and consists of nodes and connections. In ecology, biology, communication sciences, economics, geography, history, knowledge science, organizational studies, social psychology, development studies, and sociolinguistics have also developed a substantial following and are now widely used as a consumer tool (Hansson, 1994). Social Network Analysis includes social system architecture, human behavior simulation, visualization of social networks, analyzing social networks’ evolution, and graph <\/p>\n
question and mining. Online social media networks and social media research have recently become popular. The vastness of Big Data is one of the biggest challenges of Social Networking Analytics (Rathore &\u00a0Ilavarasan, 2017). The study of a network composed of millions or billions of associated objects is typically expensive in computing. Two hot science frontiers, social computing, and cloud computing favor Social Network Analysis to a certain point.<\/p>\n
Big data analytics and application for logistics and supply chain management<\/p>\n
It is clear that the challenge associated with handling large amounts of data in data science and big data analytics is the difficulty in data acquisition (Dremel et al., 2017).\u00a0 Data is the main ingredient in any analytical task. The input of any analytical task is data, and the output is information and knowledge that helps companies make data-driven decisions. Thus, data collection is one of the important steps in big data analytics. However, the acquisition of these data has its own challenges. One of the challenges of acquiring data s that data is in different formats (Dai et al., 2019). <\/p>\n
This means that different techniques will be used to acquire data from different systems. Besides, different systems are created so that they collect different types of data from different functional areas of a given company. The collected data has to be transmitted effectively to the necessary users in the company. Since the data is in large volumes, it is difficult to transmit these data to different storage infrastructures (Dai et al., 2019). One of the reasons why it is difficult to transmit large amounts of data is the amount of bandwidth required to transmit large amounts of data. The other thing that would make the processing of big data to be difficult is the redundancy nature of data. In most companies, data is collected from multiple systems. Thus, data can be redundant, which could lead to inconsistency. This is a big challenge that most companies have been trying to solve.<\/p>\n
Data-driven smart manufacturing<\/p>\n
Tao et al. (2018) posit that data privacy is a very critical issue when handling large amounts of data. Customers are always confident that when they provide their data to any company, the companies will handle their data well. However, when dealing with big data, it is difficult to ensure there is privacy of data. Take, for example, in data mining. There has been a lot of security issues surrounding data mining. This is because companies always sell their data to third parties. When the party is carrying out data extraction, the identity of the owners can be easily identified. <\/p>\n
Thus, data mining professional has an obligation to ensuring that the data is safe. Data mining professionals can easily achieve this by carrying out data de-identification. Data de-identification is the process of ensuring any piece of information, whether as a stand-alone or as a combination of data, does not lead to the owner’s identification. Since most data contains critical information about individuals, it is very important that the companies safeguard them. A lot of companies have been sued for mishandling the data of their customers and clients. Rules and regulations are being created to ensure that companies are more careful with the data that they have in their systems. Data security and privacy are very critical concepts when it comes to dealing with big data. However, most companies are more concerned with getting insights and information from data that they forget to ensure that the data’s privacy is important, and they end up compromising on the issue.<\/p>\n
Data science, predictive analytics, and big data<\/p>\n
When a company is trying to implement big data technologies, they should create a good strategy that should entail how they will ensure that the data is secure. Big data is not a new concept in big organizations. However, even small companies are implementing the big data technologies. This is because big data and data science have been shown to be very crucial in any organization. In addition, the use of big data technologies and data science has been shown to reduce the cost of operations and improve the efficiency of the overall production. Data privacy and security have a lot of requirements that companies are supposed to abide by. One of the reasons that there are privacy and security issues in big data analytics and data science is that big data technologies have been integrated with cloud computing technologies. Some of the cloud providers are very irresponsible, which could lead to data breaches. <\/p>\n
The other reason why there might be a security issue in big data analytics is that the applications designed to ensure that the data is secure cannot handle big data. Considering that one of the characteristics of big data is volume, some applications cannot handle large amounts of data which could easily lead to data breaches. Big data is also very dynamic as compared to other types of data that are static. These security applications are not able to handle the dynamic nature of big data, which could result in security and privacy issues. This means that privacy and security of big data and data science cannot be managed only by normal checks. Sophisticated methods which support big data should be used. Various prepositions have been made to ensure that there is privacy and security in big data and data science.<\/p>\n
Real-time big data processing<\/p>\n
Big data analytics in real-time ensures that big data is analyzed as it arrives. A market customer gets purchasable insights without extending a decision-making duration, or an analytical device activates an event or a warning. The real-time computing engine can be built to either push or pull data as far as the data input is concerned. A push alternative with a constant influx of high-volume data is the most common example, which is also known as streaming. The real-time processing engine, however, is not always able to absorb streaming content. Conversely, by asking if any new data has come, it may be configured to retrieve data. The time between these questions depends on the organization’s needs, ranging from milliseconds to hours.<\/p>\n
“Traditional” Big Data processing methods involving Map Reduce jobs to batch process data are not entirely suitable for real-time use cases involving non-deterministic accessibility data processing (Samosir et al., 2016). A 2013 industry study on the usage of big data technologies by European organizations reveals that over 70 percent of respondents indicate a need for real-time computing. Enabled by DSPS technologies, real-time data analysis enables data to be analyzed as soon as it is made available, without the need for a storage facility. The Storm project is a very prominent DSPS technology built independently of Hadoop, gaining considerable success and growth in its customer base. Spark is another common distributed computing system for big data, providing both real-time data processing and more conventional processing in batch mode, running on top of Hadoop YARN (Shreya et al., 2010).<\/p>\n
Web analytics, social media, and mobile apps <\/p>\n
Web Analytics Extracts data only from the Company platforms. In general, Web Analytics informs companies about your traffic rate, referral sources, bounce rate, and user behavior on companies’ websites. Web analytics use this data to improve user experience and conversion rate. The application of data analytics to optimize internet marketing dates back to creating the first web analytics programs in the 1990s. However, new analysis shows that many organizations struggle to use best practices in core web analytics and thus do not achieve the potential return from web analytics. While web analytics services are well known, web analytics technology is still not used as commonly as intended to influence marketing positively. The level of acceptance of enterprises’ instruments is strong, but their use remains relatively poor (Jayaram et al., 2015). <\/p>\n
Barriers to Web-Analytics’ effective growth in industries are mainly because most marketers lack the skill set to analyze data using Web-Analytics. Not enough advertisers have the skill sets to analyze the volumes of online data available. To understand whether the figures have gone up or down, many marketers use Google Analytics without going beneath the info and knowing why it allows for more to be done with it. Progress relies on relies on reducing already established digital marketing inefficiencies rather than running off reports as information demands occur. The ones who better show the importance of analytics can speak to senior executives, consider their challenges, and then offer analytics answers. The senior managers would never have dreamed about asking the web analyst to assist them without this.<\/p>\n
It is believed that Web Analytics needs a close analysis of how commercial value is created now and, in the future, to establish a digital marketing optimization approach that enhances the contribution of web analytics to organizations. An important starting point is to analyze how the future value of implementing web analytics relates to business success and then compare it with existing capabilities and generated value (Chaffey & Patron, 2012).<\/p>\n
Social media analytics captures data from social media networking platforms and lets organizations grasp customer sentiment, consumers’ behaviors, build rich audience profiles, and, most critically, build successful business strategies. It is commonly used to find the customer’s emotions, sentiments, and feedback after using a product or an application.<\/p>\n
Social networking data has a vast amount of knowledge. Companies paid market research firms to interview clients in decades past and hold focus tests to get the kind of information that customers are now voluntarily sharing to public social media sites. <\/p>\n
Almost all companies now use social media to promote their companies and get feedback on their performance. This helps companies to retrieve data on how the customers view their brand, what sort of goods people want and hate, and where sales go in general. Without using less accurate surveys and focus groups, social media analytics helps companies to measure all this.<\/p>\n
However, it is increasingly understood that seeking ways to interpret what social media analytics teaches us about social life; such learning carries with it new analytical challenges. Social networking offers a type of user-generated content that can be unsolicited, unscripted, and multi-modally conveyed via music or texts. Therefore, to make them susceptible to interpretation and recognize the kind of research problems that such data might properly answer, it is important to acknowledge the difficulties this data has for researchers (Brooker et al., 2016).<\/p>\n
Mobile analytics gathers information from visits to mobile applications, websites, and web apps to recognize specific users, monitor their movements, document their actions, and report on their success. Web analytics is used to maximize conversions, comparable to conventional web analytics, and is the secret to developing world-class digital apps. <\/p>\n
Mobile analytics offers enterprises unique access to smartphone users’ otherwise hidden lives. Analytics generally comes in applications that incorporate data processing, storage, and interpretation into organizations’ current websites and apps. This knowledge is important for marketing, distribution, and stock management departments to make more educated decisions. Businesses are left flying blind without a mobile analytics solution. They cannot know what users are interacting with, who those users are, what attracts them to the website or app, and whether they are leaving. <\/p>\n
The use of mobile exceeded laptop and desktop use in 2015, whereas smartphones are quickly becoming the internet’s preferred portal for users. Users spend 70 percent of their video usage and screen time on mobile devices. This is a massive opportunity for firms to meet their clients, but it is still a heavily saturated sector. The big smartphone app stores, millions of online applications, and more than a billion websites have more than 6.5 million apps. In creating mobile apps that stand out, companies use mobile analytics platforms to achieve a strategic advantage. For both mobile and desktop computers, most current product analytics platforms monitor consumers. Users have less real estate on the smartphone phone and connect by touching, swiping, and holding. As a result, with fewer navigation tools, smartphone applications and website sites are easier to use. Users have bigger displays on a laptop and communicate by clicking, double-clicking, and using key commands. More connections, more content, wider menus, and more links per tab are normally included in desktop monitoring. This user disparity can be compensated for by a strong mobile analytics platform to have one unified, centralized dashboard that identifies specific users and their activities through platforms.<\/p>\n
Consumers will expect business applications running on mobile devices to have expanded functionality and capabilities as mobile devices’ usage continues to expand. Features associated with mobile devices can affect the functions and capabilities of applications operating on mobile computing platforms. However, customers can require apps operating on mobile devices to have the same or equivalent capabilities as apps running on large or immovable business computing platforms. On a mobile computer with an interface for a remote link to a backend server, a mobile analytics engine applies a business analytics model (Shanks & Bekmamedova 2012). The mobile analytics engine uses metadata models to automatically adjust query or report content under resource restrictions or constraints of the mobile device for customized query or report execution (Eberlein & Said, 2015).<\/p>\n
The mobile analytics engine supplies the presentation layer of mobile devices with analytics and reporting results. The mobile analytics engine uses report and analytics metadata objects identified in a mobile metadata repository to present a generic interface to serve diverse mobile business analytics applications. <\/p>\n
Role of data scientists on application development teams<\/p>\n
Creating and running applications generates vast volumes of raw data about the method of creation and consumer use, which can be transformed with professional data analysts’ aid into actionable knowledge. Unfortunately, it has not been easy to come across data scientists with the computational and information engineering expertise to analyze these broad data sets; only recently have software firms begun to build software-oriented data analytics competencies (Kim et al., 2016). <\/p>\n
Different working styles of data scientists are as follows:<\/p>\n
Insight suppliers who collaborate with engineers to gather the knowledge required to inform managers’ choices.<\/p>\n
Modeling experts who use their knowledge of machine learning to build predictive models.<\/p>\n
Platform designers, who construct information systems, balance the problems of engineering and data processing.<\/p>\n
Polymaths, who do all the tasks of data science themselves.<\/p>\n
Team Leads, who manage data scientists’ departments and share best practices.<\/p>\n
Public policy considerations for data-driven innovation<\/p>\n
Hemerly (2013) argues that a lot of companies and projects have worked towards ensuring that they are able to solve the privacy and security issues in big data analytics and data science. OSLO analytics is one of the projects that have been able, although not fully, to ensure there is data privacy and security. Oslo is a company that carries out big data analysis and creates machine learning models that help to understand and create awareness about security incidents. Oslo analytics work together with national organizations, international organizations, and security companies in Norway to ensure that cyber security is maintained (Innes &\u00a0Booher, 2018).\u00a0 Oslo analytics carries out an alloy of research that is based on personal information. <\/p>\n
Thus, one of the obligations that Oslo analytics has is to ensure that the data they have is secured. It is of great importance for Oslo analytics to use anonymous data. Anonymous data is data that cannot identify the identity of an individual either directly from example by use of names or indirectly. Since Oslo is an analytics is a project that whose main raw material is data, privacy and security of the data is o crucial importance. Oslo analytics has worked hard to ensure that security and privacy are being maintained. One of the ways that Oslo analytics is able to ensure that there is privacy and security of data is by ensuring that the storage of data is secured and strict measures are put to limit accessibility. Oslo analytics store their data on very secure servers.<\/p>\n
Applications, prospects and challenges<\/p>\n
Vassakis et al. (2018) emphasize that app development companies use sophisticated measures to ensure that the researchers working on the project are the only ones who can access the data; in addition, since various researchers work on a different topic. Every researcher is able to access the data that is important in order to solve the problem at hand. Oslo use the method of least privileges to ensure that there is a restriction of access to the servers. In addition, the servers can only be accessed when the researcher is inside the organization. This helps to ensure that a researcher cannot misuse the data which could bring about privacy and security issues. In addition, it also ensures that no one outside the organizations, other than the researchers can access to the severs. <\/p>\n
The Oslo analytics also installed a firewall to ensure that only allowed connections are able to access the server. This is important to ensure that any negative traffic is eliminated. In addition, Oslo also restricted the installing of other programming software. This is important to ensure that no data is leaked outside the organization, which can compromise the privacy of data. The last strategy that Oslo analytics put to ensuring that access is restricted is that the accounts that they have created from their researcher will not be activated when the project is completed. Thus no data can be able to go beyond the premises of the organization.<\/p>\n
Creating a data-driven organization<\/p>\n
App developments companies ensure that there is data privacy and security and are able to balance between productivity, security, and availability of data. Organizations are using data to increase their productivity. They are trying to improve their productivity at the expense of the security of their data. This is one of the practices that has increased the number of breaches that have been reported in organizations. The major objective of Oslo analytics is to analyze the data at hand and provide knowledge about security.<\/p>\n
Reproducible research demands that the analysis that is done by any organization should be published to ensure that other parties can assess the findings of the analysis. Thus, this data is very prone to attack, and any malicious party can try to steal the dataset. The network that is connected to the anonymized dataset can be used to detect any malicious party that is trying to steal the dataset. Analytic companies use end point security to help them to handle sysmon data. One of the data that Oslo analytics uses to ensure that there is privacy and security of data is Sysmon logs. Symon is a device that uses to record the logs to the server systems and pother domain controls. Thus they can be able to detect any malicious logs that happen on any particular day. <\/p>\n
The Sysmon provides one of the most trusted data that can be used to detect the activity of any kind of attack. The Sysmon system is able to collect very sensitive identifiers that Oslo is much interested in. Organizations should implement a system that is able to track the interactions between different users and their systems. Thus, it becomes very easy for organizations to detect any incoming traffic that can be suspicious.<\/p>\n
Toward data-driven requirements engineering<\/p>\n
Maalej et al. (2015) emphasize that surveys are used to outline the various methods that are available which organizations, companies, and projects can utilize in order to ensure that there is data privacy and security. The first technique that the researcher recommends is the use of cryptographic approaches in handling big data analytics. Cryptography offers a wide range of methods that can be used to ensure that there is security in big data analytics. Organizations that have implemented the use of big data analytics in most cases have implemented the use of cloud computing services. <\/p>\n
The first cryptography method is the use of homomorphic encryption. This method ensures that data functions can be written without decrypting the data first. This is method has been proven to be among the best methods in ensuring that sensitive data is protected. The second cryptographic method is verifiable computation. In verifiable computation cloud nodes are not allowed to perform any data integrity.<\/p>\n
Thus the responsibility of data surety and privacy is entirely on the client. This helps to ensure that no third party has access to the data of the organization. The second method that is used to ensure that there is data privacy and security is ensuring that the data storages are secure. In addition the sharing scheme of the tenants in the cloud should also be secured. Organizations should use techniques such as the least privileges technique to ensure that only the authorized persons have access to the data. Most companies are migrating to the cloud. In some cases, the data that is stored in the cloud is not always secure. Data always has various elements that are mapped directly to the data.<\/p>\n
The opportunity and challenge for IS research<\/p>\n
The cloud providers should always ensure that these elements are secured; otherwise ,other tenants of the clouds can easily access the data while still in the cloud. This method helps clients to divide their data among various cloud providers and use trap doors to keep the data from being accessed by unauthorized parties. The third technique of the perturbation method. This technique involves data masking. An individual can be able to fully use the masked data if the characteristics of the statistical requirements are the same as the original data. Hence the data can be hidden for the purposes of security to ensure that the only person who accesses the data are the only ones who require the data. <\/p>\n
There are various techniques that can be used to mask data. These includes matrix masking and post-randomization method. The perturbation method is said to have a great impact on the privacy and security of data. One of the major reasons is that the method is very easy to implement and also takes a very short amount of time to implement. In additions, this is the most reliable method among all the other techniques for security and privacy maintenance. Since perturbation is based on a statistical analysis of data, it is one of the best approaches that are used to ensure data security and privacy in big data analytics.<\/p>\n
The first researcher used a case study in order to show how privacy and security of data can be achieved in any project and organization. The organization should be able to understand that privacy is so critical in big data analytics. The second researcher used the survey method to outline the potential solutions to data privacy and security. There are many techniques that should be used to ensure that there is data privacy and security. A company should choose a method that aligns with its objectives and one which will help it to realize the full potentials of big data analytics and data science. In order to ensure that the efficiency of the method is realized, the time factor should be considered. An organization should also choose a simple technique that its employees will enable to comply with. The last thing that employees need is anything that will make their daily tasks more complex. <\/p>\n
Thus, the technique should not be complex. The challenges that face big data analytics must be addressed in order for an organization to realize its potentials. When it pertains to big data and data science, privacy and security are very important concepts. Each organization should work towards ensuring that it has a working method or technique that helps to ensure that there is data privacy. This paper has discussed some of the challenges that organizations face as they are trying to handle big data, the biggest challenges being privacy and security issues. Privacy and security are such a critical concepts that every organization should be very keen on when dealing with big data analytics. There are various techniques that organizations can use to ensure that there is privacy and security in big data analytics. One of the methods is the use of cryptography to ensure that the data deployed in the cloud is safe.<\/p>\n
The role of data-driven e-government in realizing the sustainable development goals in developing economies<\/p>\n
Everyday computer forensic scientists carry thousands of data acquisition activities. The acquisition is majorly for evidence reasons. Whether leading or confirming a specific allegation. This puts their job in the frontline of justice delivery. Several methods such as logical, physical, static or sparse with corresponding software can be used. <\/p>\n
These methods are chosen depending on the desired form of data, time available, and media available and size of data being acquired.<\/p>\n
Logical acquisition as a method of data acquisition for the 2GB data would capture the specific files of interest in the disk. This method would save time for the user as they wouldn\u2019t have to dig through a bulk. The particular software to be used for the logical acquisition would be Micro Systemation XRY on a clone cellphone tool as the media. <\/p>\n
Sparse investment, on the other hand, would involve the capture of data on the disk plus any other fragmented data such as data deleted, therefore bringing out a bulk. The software to use for the sparse acquisition would be X-Way\u2019s Imager which would run directly from a flash pen without the need to. The static addition would work best for data that is write-protected, which means it can’t be altered. The software to use for this method would be Utimaco Software Safeguard Easy which depending on user preference can run on laptops, desktops or PDA.<\/p>\n
In the CRM application system, maintenance, storage and updates are performed by the service provider. However, the on-premise CRM system is supervised by the company’s IT department. A cloud-based system is easier to use and provides immediate access to vital information throughout the sales cycle. A cloud-based system is cheaper, with reduced costs upfront. However, initial investments for traditional methods are expensive. Traditional CRM systems can only be accessed during office hours from an in-house system. On the other hand, cloud-based CRM systems can be accessed anytime at any place hence improving the efficiency of the organization. <\/p>\n
The integration of cloud-based systems is usually low, therefore minimal capabilities. The cloud-based system has a robust integration with numerous third-party items, enabling companies to perform multiple tasks in multiple tasks in multiple departments easily. The comparison illustrates the key differences that show why companies ought to have cloud-based CRM systems. One of the most significant benefits of a cloud-based CRM system is that it is accessible. The prospective user only requires an internet connection. This accessibility is possible because the server is on the cloud and not in a specific geographical location. As such, employees can perform their tasks within or outside the company premises. Furthermore, international teams in different nations and time zones can access the CSRM system when the need arises. <\/p>\n
Challenges in mobile app development<\/p>\n
With leading-edge innovations, technology-oriented start-ups are rising all over the world. Since consumers commonly use smartphones, more development firms spend more resources on developing mobile applications to take advantage of the rising demand for apps.<\/p>\n
In recent years, the field of mobile apps has seen tremendous growth. Therefore, many entrepreneurs are planning to develop smartphone applications as the top of their industry plans for their company maintenance(Hammershoj et al., 2010). It’s really important to consider the problems and build a strategic solution to solve them.<\/p>\n
For the last couple of years, millions of smartphone apps have been uploaded to Apple and Android app stores and have become a significant metric for a profitable market. I’m going to clarify and illustrate the key obstacles to mobile app growth here in this post.<\/p>\n
Validation and Assessment of customer requirements:<\/p>\n
One of the greatest obstacles in evaluating the course and finding out what to build is consumer needs evaluation. There is tremendous competition in the smartphone app industry market, and it is a challenging challenge to get the stand and be heard by consumers. The creators of applications still strive to build creative software that consumers like. This can be done with the help of Data Analytics, as it requires customer data.<\/p>\n
Defining the demand for target\/competitive:<\/p>\n
The industry competition is too aggressive when thousands of smartphone users seek to approach the crowd with creative stuff to stand out. It is tricky to reach out to shorter-span customers on the market. The app creators should realize which innovation can draw consumers with the highest values over a longer period. To find out how they can keep their customers more involved and interactive, they can study more thoroughly. To ensure that a loyal user base is established, the software developers can take on the difficult task of innovating the app with a high quotient. This would affect outlining the target market from a commercial viewpoint, and with a definite market, the commodity will get great monetization.<\/p>\n
Resource management:<\/p>\n
You may have a great and creative app concept, but if you are not trying to turn it into marketable money, it will go in vain. For an app, the cost of production often depends on the quality of the app. It would be tough for certain founders or developers to monetize the app and handle the resources and financing.<\/p>\n
Again, you will involve getting a loan or joint venture with your project with others to collect funds if you don’t find a suitable channel to invest in your app concept. It can be a difficult challenge to handle the funds successfully to deliver a profitable return on investment.<\/p>\n
Choosing technologies for development:<\/p>\n
A determining consideration for the developers would be selecting a central, constructive, and acceptable development technology for an app. This gives you an idea of whether a native, hybrid, or cross-platform app should be made.<\/p>\n
Lots of creative innovations are spread throughout the industry. But it can be hard to develop a good understanding of the benefits and drawbacks that each technology that serves your platform provides. Choosing an obsolete technology will lead to more investment, poor performance of the app and a greater customer experience. It’s always a daunting challenge to create an interface that is scalable and adaptable according to the customer’s goal.<\/p>\n
Security<\/p>\n
In-app security concerns would be a problematic challenge for developers. The app should be free from malware concerns; otherwise, it will create disintegration of software or hardware that would take more time and resources to fix those problems.<\/p>\n
Ensuring protection on smartphones can be a difficult challenge to achieve maximum safety levels with the prevalent operating systems and software models. For software developers, creating a link with end-users is a demanding task. The full effort and money will go in vain if the smartphone app fails to communicate with the end-users.<\/p>\n
App Marketing and promotion:<\/p>\n
When you create an app, promotion is a critical thing. Many developers don’t place any focus on ads. For mobile developers at a different stage, this is difficult and needs market savvy to thrive. An app with a nice interface, graphics, and features can be created, but if they don’t do better marketing, they won’t get the investment returns.<\/p>\n
Big data: the management revolution<\/p>\n
Both Facebook and we chat on social media platforms. Facebook is based in the United States of America, while WeChat is based in China. However, Facebook and WeChat offer a wide variety of services, and they differ in their services. WeChat offers voice, textual, and broadcast messaging services. Also, WeChat offers video games and video calling services for both conference and single. Besides, when two we chat users are physically close, they can exchange their contacts through Bluetooth technology. On the other side, Facebook is majorly a web-based application and has mobile apps for android, iOS Windows, and Tizen-powered mobile devices. Also, Facebook massager users the default Facebook account. Both WeChat and Facebook have comprehensive functionalities depending on the targeted geographical location. For instance, if you are targeting more on the Chinese, WeChat will offer more services, and if you are targeting more on Americans, Facebook will offer a wide variety of services.<\/p>\n
A taxonomy of data-driven business models used by start-up firms<\/p>\n
Cloud-based systems minimize human error. Data is stored in one location, which implies that data reporting and analysis are improved. Reports are generated automatically, thus resulting in higher productivity by consuming less time. There is also the option to personalize dashboards to instantly access needed information like client data, performance reviews and sales targets that may reveal opportunities for the corporation. Improved reporting facilitates informed decision- making that may lead to retention of consumer loyalty and more profitability (Mintzberg & Westley, 2001). Consumers are an important part of the business, and it is not worthwhile to risk that relationship by using traditional based CRM systems. <\/p>\n
Management Review<\/p>\n
Investing in a cloud-based system encourages service provision and customer satisfaction. In summary, cloud-based CRM systems are better than traditional on-premise systems. Conventional systems require high initial capital investment, are hard to integrate, costly to maintain, among other challenges. On the other hand, cloud-based systems are affordable, accessible and limit human error. Traditionally, firms were required to purchase the hardware and software, cater for upgrades and maintenance costs. There were also hidden costs that made it difficult for small and medium enterprises to afford these systems. However, cloud-based systems are useful for business leaders who have a constrained budget, inadequate resources or do not have an IT team in the house. Numerous cloud-based systems may require a monthly fee while others a flat rate. They also come with upgrades and maintenance, which implies that money that could be used to access these services, is saved for other purposes.<\/p>\n
Data mining with big data<\/p>\n
Algorithmic decision-making by the China government poses a challenge to the rule of law. However, we cannot ignore the benefits that algorithmic systems would bring in the delivery of services. China needs to develop the right environment for the adoption and development of algorithmic systems. Therefore, the algorithmic system needs to be checked to ensure that there is no bias based on class, gender, or sexuality. The law must avoid being rendered irrelevant and come up with ways in which algorithmic systems can be used to enhance human dignity rather than remove it. If the algorithmic system is checked, it may end up finding common ground with the rule of law. <\/p>\n
Harness the power of big data<\/p>\n
Chatbots are programs that necessitate human interactions through voice interface or typed messages. Chatbots have developed the inclusion of artificial intelligence systems that interact with humans through voice or text interface and can be compared to personal assistants (Duan et al., 2019). Chatbots are found in significant applications such as Slack, Facebook, and text messages. However, despite that they have developed artificial intelligence, chatbots are inferior for chatting (Frantz, 2003). Chatbots have evolved disruptive technology for the communication sector. Chatbots can interpret the meaning of what was typed or what was said even after the communication was deleted. Also, the intelligence of chatbots is limited compared to what artificial intelligence is supposed to offer to a client (Grabo\u015b, 2004) In most cases, chatbots post irrelevant information hence wasting time for the client, and this makes them inferior for chatting. <\/p>\n
Data privacy and security<\/p>\n
The organizations should build a successful plan as they want to introduce big data technology, including how to ensure that the data is protected. In large companies, big data isn’t a novel phenomenon. Big data innovations, though, are being introduced by even small businesses. Also, it has been shown that the use of big data technology and data science decreases operational costs and increases the quality of total production. There are several standards to be fulfilled by enterprises for data protection and safety. Big data analytics and data science have privacy and security concerns because big data applications have been combined into cloud computing technologies. Any of the cloud providers, which may lead to data leaks, are quite reckless.<\/p>\n
Big data analytics can have security issues because big data cannot be managed by software developed to ensure that the data is secure. Provided that scale is one of the big data features, some systems do not accommodate massive volumes of data that could easily lead to data breaches. Compared to other types of data that are static, big data is often very dynamic. The complex existence of big data, which may lead to security and privacy concerns, cannot be addressed by these security applications. This suggests that it is impossible to handle the privacy and protection of big data and data science only by usual controls. It is important to use sophisticated techniques that help big data. Numerous prepositions have been made to ensure privacy and confidentiality in big data and data science.<\/p>\n
A data-driven organization<\/p>\n
Companies that develop apps ensure that data protection and confidentiality are accessible and can balance efficiency, reliability, and data access. Companies use data to improve their efficiency. At the cost of the protection of their files, they are trying to increase their competitiveness. This is one of the behaviors that has raised the number of corporate violations reported.<\/p>\n
To achieve strategic advantage and optimize their decision-making by being data-driven organizations, organizations are searching for opportunities to leverage big data and integrate the change that big data brings into their business strategies. Despite the possible advantages of being data-driven, the number of organizations that use it effectively and turn it into data-driven organizations effectively remains limited. Most of the literature’s focus was technology-oriented, with little attention given to the operational problems involved (Mesquita, 2019).<\/p>\n
Increasingly, software development firms seek to become data-driven by attempting to play with the goods used by their consumers on an ongoing basis. They rarely excel in designing and implementing the technique, although they are familiar with the A\/B research technology’s competitive edge. This article presents the evolutionary method of going from ad-hoc consumer data research to continuous managed testing at scale focused on an extensive and collective case study in a major software-intense organization with a deeply established experimentation community (Fabijan et al., 2017).<\/p>\n
Challenges in mobile app development<\/p>\n
With leading-edge innovations, technology-oriented start-ups are rising all over the world. Since consumers commonly use smartphones, more development firms spend more resources on developing mobile applications to take advantage of the rising demand for apps.<\/p>\n
In recent years, the field of mobile apps has seen tremendous growth. Therefore, many entrepreneurs are planning to develop smartphone applications as the top of their industry plans for their company maintenance (Hammershoj et al., 2010). It’s really important to consider the problems and build a strategic solution to solve them.<\/p>\n
For the last couple of years, millions of smartphone apps have been uploaded to Apple and Android app stores and have become a significant metric for a profitable market. I’m going to clarify and illustrate the key obstacles to mobile app growth here in this post.<\/p>\n
Validation and Assessment of customer requirements:<\/p>\n
One of the greatest obstacles in evaluating the course and finding out what to build is consumer needs evaluation. There is tremendous competition in the smartphone app industry market, and it is a challenging challenge to get the stand and be heard by consumers. The creators of applications still strive to build creative software that consumers like. This can be done with the help of Data Analytics, as it requires customer data.<\/p>\n
Defining the demand for target\/competitive:<\/p>\n
The industry competition is too aggressive when thousands of smartphone users seek to approach the crowd with creative stuff to stand out. It is tricky to reach out to shorter-span customers on the market. The app creators should realize which innovation can draw consumers with the highest values over a longer period. To find out how they can keep their customers more involved and interactive, they can study more thoroughly. To ensure that a loyal user base is established, the software developers can take on the difficult task of innovating the app with a high quotient. This would affect outlining the target market from a commercial viewpoint, and with a definite market, the commodity will get great monetization.<\/p>\n
Resource management:<\/p>\n
You may have a great and creative app concept, but if you are not trying to turn it into marketable money, it will go in vain. For an app, the cost of production often depends on the quality of the app. It would be tough for certain founders or developers to monetize the app and handle the resources and financing.<\/p>\n
Again, you will involve getting a loan or joint venture with your project with others to collect funds if you don’t find a suitable channel to invest in your app concept. It can be a difficult challenge to handle the funds successfully to deliver a profitable return on investment.<\/p>\n
Choosing technologies for development:<\/p>\n
A determining consideration for the developers would be selecting a central, constructive, and acceptable development technology for an app. This gives you an idea of whether a native, hybrid, or cross-platform app should be made.<\/p>\n
Lots of creative innovations are spread throughout the industry. But it can be hard to develop a good understanding of the benefits and drawbacks that each technology that serves your platform provides. Choosing an obsolete technology will lead to more investment, poor performance of the app and a greater customer experience. It’s always a daunting challenge to create an interface that is scalable and adaptable according to the customer’s goal.<\/p>\n
Security<\/p>\n
In-app security concerns would be a problematic challenge for developers. The app should be free from malware concerns; otherwise, it will create disintegration of software or hardware that would take more time and resources to fix those problems.<\/p>\n
Ensuring protection on smartphones can be a difficult challenge to achieve maximum safety levels with the prevalent operating systems and software models. For software developers, creating a link with end-users is a demanding task. The full effort and money will go in vain if the smartphone app fails to communicate with the end-users.<\/p>\n
App Marketing and promotion:<\/p>\n
When you create an app, promotion is a critical thing. Many developers don’t place any focus on ads. For mobile developers at a different stage, this is difficult and needs market savvy to thrive. An app with a nice interface, graphics, and features can be created, but if they don’t do better marketing, they won’t get the investment returns.<\/p>\n
Conclusion<\/p>\n
Even though big data can superfluously be seen to be serving the same purpose, a more in-depth study by scholars such as Dhar 2013; Tiwari, 2014; Provost & Fawcett 2013 reveals that the effects of big data are reflective of organizational knowledge dissemination practices. Provost & Fawcett (2013) argues that big data analytics have led to changes in trends with decision-making for organizations developing mobile apps becoming data-driven. Differences between data-driven app development and the orthodox ones driven by speculation are almost blatant in the performance of different apps. <\/p>\n
A much narrower range of basic principles containing data science is at the heart of the vast array of mining data techniques. We must look at the algorithms, methods, and resources in common usage for data science to thrive as a profession rather than drown in the stream of public interest. We have to think about the fundamental values and ideas that underlie the processes and the systemic thinking that facilitates data-driven decision-making success. These principles in data science are general and apply very widely (Provost & Fawcett, 2013). In today’s data-oriented market climate, achievement involves thinking about how these basic ideas are related to real business problems-to think data-analytically. This is aided by computational constructs that are part of data science itself.<\/p>\n
Data science facilitates data-driven decision-making, which also allows for massive-scale decision-making automatically and focuses on “big data” storage and engineering technology. Though, data science concepts are their own and should be specifically considered and addressed so that data science can understand its potential. Provost & Fawcett (2013) claims that big data research has contributed to improvements in decision-making patterns become data-driven for companies creating smartphone applications. In the success of various applications, disparities between data-based software creation and orthodox ones driven by speculation are almost evident.<\/p>\n
Chapter 3<\/p>\n
Introduction<\/p>\n
Like any research project, the study seeks to achieve various objectives. It sets the context for analyzing the study on data science and analytics within organizations while considering what is not within the study’s scope and justifies why organizations should incorporate data management and analytics into their operations. It also establishes the existing literature using various scholars and past settings. It details the claims presented in the existing study and assesses the research techniques applied to provide a better understanding of whether the claims are valid. Such assessments allow one to separate the learned and achieved facts and those that need further studies. Additionally, the review creates a summary of existing literature to develop a way that allows for new perspectives. Thus the research provides a basis of the methodological and theoretical complexities, thus improving the application and quality of the study. The chapter outlines the study’s background and clearly understands the gaps in the study. There are discussions on how to deal with the gaps identified. The review connects different readings to create a discussion on the use of data in organizations.<\/p>\n
The objective in carrying out the quantitative study is to find the correlation between data and organizations. This type of research is descriptive since the subjects are assessed once, and the study looks at the association between organizations and data. The research deals in various logic since the data does not change and applies convergent reasoning, meaning that creating various ideas on the research problem is free-flowing and spontaneous. Like other types of quantitative research, the data is collected using structured study instruments like the Leavitt model. The findings are based on sample sizes that provide an overall picture of data-driven organizations and those that are transforming to adopt a data-driven perspective. Since it is highly reliable, the study can be repeated. The literature has various research questions that are clearly defined, and their answers are shown. Various aspects of the research have been developed prior to the collection of data. The projections shown in the research could be used to forecast future findings, generalize ideas, and assess various correlations. There are tools like the Leavitt model that being used to assess information. The research aims to categorize different factors in relation to data and organizations and create models that enable one to discuss the observations.<\/p>\n
Some of the things that have been considered when describing the findings of the study include the data gathered, their statistical approach, and the relevant findings that relate to the research questions being assessed. There are no unanticipated occurrences that took place during data collection (Amankwah-Amoah &\u00a0Adomako, 2019).\u00a0 The actual assessment is different from the planned analysis, and this is attributed to the lack of information on areas like data description and the difference in how they are collected and disseminated. Even though there are variations, the difference is insignificant, and the research handles missing data in ways that do not affect the assessment’s validity.<\/p>\n
Some assumptions include theories derived from studies that indicate that people’s behaviors are incompatible with normative concepts. It concentrates on how and why individuals act in certain ways without any influence. Another assumption is that real-life decisions can be rational and non-rational (Bell et al., 1988). Also, considerations have been made to assume that those tasked with making decisions always understand the issues they face; thus, they can express the issue as an efficient, or effective issue, and they possess the right resources and information to develop solutions. This is not true since those tasked with making decisions usually have a vague depiction of the issue. The issues can instead be expressed when searching for appropriate compromises and solutions to various issues are always affected by time and the resources that are readily available (Tsouki\u00e0s, 2008). Additionally, good decision-makers should be able to make decisions while applying any decision-making procedure.<\/p>\n
Additionally, the theory implies that those tasked with decision-making select among known and fixed options, and both have known impacts. However, this is inaccurate since cognition and human perception are reliant on the decision-maker and the surrounding. Contrarily, when considering the decision procedures, alternatives are not provided but should be searched. Determining the impacts is a difficult endeavor since the information possessed by decision-makers on their surroundings is less compared to actual estimations of the information within their surroundings. The last assumption is that analytics and big data lead to better decisions, but the assumption is simple and is not based on research (Bartkus et al., 2018). Chain perspectives should be done to better understand the various factors that impact data-driven decision-making and their correlation and autonomy with other activities and procedures.<\/p>\n
Research Paradigm (quantitative)<\/p>\n
The research paradigm indicated by the research is pragmatism since it offers a guiding structure based on practicality. This is appropriate for the assessment since performance analysis is applied in organizational settings. It allows the research to create procedures as organizational practices and create a multi-faceted and dynamic comprehension of the practices. By indicating the method’s suitability for research on data and organizations, the research adds to a growing discussion that supports the use of data in organizations. There are suggestions that the research may be particularly reliable in guiding studies to offer organizational improvements by bolstering policy and practice while staying true to the difficulties of carrying out research, defining various concepts in the research help in precise data gathering and strengthening the quality and depth of the research. The concepts of actionable awareness founded the study in the experience of the organizations assessed. At the initial phase, the concept allowed the study to unwrap the research puzzle and recognize factors of the problem that are relevant. As the research progresses, the concepts help in determining the methods used and the research questions. By highlighting the concept of actionable understanding and the study procedure, one is able to identify the study’s findings that are relevant and can contribute to the concept and practice of organizational data management and analytics.<\/p>\n
The research maintains that for data management to grow within organizations, we should think beyond the techniques, algorithms, and tools that are commonly used. The human ability allows for the consideration of core concepts that define the techniques and also the methodological thinking that nurtures success in data-instigated decision-making. The ideas are basic and applicable. Improved performances can be sustained in data-oriented organizations if they can think of the key factors and apply them to specific business issues. This is assisted by conceptual structures from data science. For instance, the computerized reading of patterns from data is a procedure with clearly spelled phases (Bean &\u00a0Davenport, 2019).\u00a0 Comprehending the procedures and the phases assists in structuring problem-solving mechanisms, making it systematic with reduced chances of error. Strong grounds indicate that organizational performance can be enhanced significantly through data-driven decision-making, analytics, and big data. Data science emphasizes data-driven decision-making and sometimes enables the automated decision-making of decisions on a large scale, mainly due to engineering and big data usage. It is important that research materials discuss both (data science and big data) distinctively so each can retain its meaning.<\/p>\n
The research seeks to answer the research questions by examining the topic(s) to create a decision theory that accounts for the abilities of data-instigated decision-making by incorporating the decision-making factors with contemporary developments in data analytics and big data. The theory will act as an epistemological ground for promoting the efforts of data-driven decision-making, giving reasons that stretch past the theories, thus allowing for future studies (Hansson, 2011). Comprehending the interrelation between data, humans, and organizations influenced the research design approaches used in the study help in solving various issues. The concept is important in guiding various research strategies that allow for the recognition of diverse literature that is rich in information and have a high chance of providing helpful insight and ensuring the strategies consider multiple perspectives. Also, by mapping out some of the likely impacts of the use of data in organizational decision-making, we will acquire a detailed comprehension of the situations being investigated.<\/p>\n
Also, the understanding of the study as an experiential procedure improved various research abilities. In every phase of the study procedure, various ideas are interrogated and evaluated in terms of their functioning. The concepts allow for the combination of different perspectives within the study and promote the use of an inclusive process that involves different kinds of literature that one can relate to and apply their actions in an authentic context. The paradigm allows for flexibility and is adaptive through the research procedure since there are informed methods to improve the study projects’ effectiveness and value. The implosion in the creation of big data along with the creation of new concepts is making many consider that there may be a data revolution approaching, and this may have various impacts on how knowledge is created, businesses are done, and governance is executed (Anderson, 2008).<\/p>\n
There is also the possibility of a new research paradigm across the different disciplines identified in the research. As indicated by various descriptions, a paradigm is made up of various acceptable methods of examining concepts and integrating knowledge that is unique to a specific group of studies. It creates new ways of reasoning that challenge the approaches and theories that are deemed acceptable. Unlike previous assumptions that paradigm changes happen since the existing literature cannot account for certain concepts or answer major questions, thus the need to create new ideas, these transitions are based on developments in analytical methods and advances in new types of data.<\/p>\n
There are suggestions that a new paradigm is instigated by the increasing availability of analytics and big data (Boyd &\u00a0Crawford, 2012). Various arguments have been critical of this due to the lack of substantial evidence indicating how paradigms operate, especially in various social sciences that have various sets of theoretical methods used, even though in various domains, like sciences, there have been increased unity on the manner in which science is done. Even though the idea of paradigms is questionable, it shows uniformity in developing the discussions that touch on the creation of big data and their impacts since many of the claims in relation to knowledge production indicate that organizations that transform into data-driven entities result in transitions that can always lead to new paradigms.<\/p>\n
Just like in societies and businesses, data is increasing in volume, variety, velocity and calls for new methods of deriving value. The systematic removal of information from data has been eliciting a lot of attention. It is stated that data science is a new paradigm. Various disciplines continue to discuss their epistemological challenges, assumptions, and opportunities. Also, there are concerns about whether it is a paradigm, empiricism, or just an extension of existing ones with new approaches and tools for inquiry.<\/p>\n
The concepts in the information system are also distinct. Data science is seen as a methodology, paradigm, or phenomenon of research. The methods basically discuss the challenges and opportunities of utilizing data management in organizations. This results in innovative ideas, though the major factors are often ignored. The research, like other materials on data science and data management, provides different views on how they are applied in various institutions. Additionally, it views data science as a way of creating theoretical contributions in various fields.<\/p>\n
Data science is considered a paradigm that comes with its unique theoretical assumptions. Innovative paradigms result from the following reasons; a measurement revolution, developments in the creation, collection, analysis, and management of data, hindrances in current paradigms’ capabilities in pushing for theory development and knowledge discovery. Whether the above-mentioned reasons instigate the paradigms changes, there is the idea that a new paradigm is emerging in science. The paradigm is linked to a data revolution and the possibilities it brings. Though intense discussions are ongoing on its nature, people should be cautious not to encounter the same difficulties as empiricism that are parallel to today’s discussions on data.<\/p>\n
Each paradigm offers a different perspective on what defines a valid theory and its roles. This creates a lot of emphasis on various choices instead of study objects while increasing the risk of creating irrelevant knowledge that has no value. As we look at the methodological problems when looking at data science, the research reviews the common goals of data science which is meant to develop an understanding of the study objects and validity. These goals can be looked at in terms of commonalities instead of the variations among the knowledge types autonomous of their respective paradigms.<\/p>\n
However, the perspective that the new paradigms are taking is debatable. The remainder of the research analyzes the development of this paradigm by exploring the extent to which the transformation to data-driven entities results in other paradigms in various fields and transforming study practices. This type of paradigm is data-intensive and a rapid extension of the existing scientific approaches. It also brings in a period of empiricism in which the amount of data, followed by methods that can indicate facts, allows data to be used independently. This type of view has acquired confidence outside learning institutions and grown into business entities, and some of its concepts have gained ground in data science and analytics. It is important to critically examine various approaches while considering the motivations and aspirations of various organizations. This includes using data analytics to find new markets, products, and opportunities.<\/p>\n
There is the perception that the creation of big data and new data analytics provide opportunities for restructuring in organizations, and such restructuring would happen across all the departments. Data analytics and big data allow new approaches for generating and analyzing information, thus making it possible to ask and answer questions using new techniques that are reliant on machine learning and visualization.<\/p>\n
Access to big data has resulted in creating a paradigm based on data-intensive exploration that examines various scientific approaches. Even though evidence shows that big data is disruptive, empiricism allows the data to speak for itself without having to incorporate theories and data-driven abilities that rapidly change the current methods by incorporating various factors of induction, abduction, and deduction. Due to the weakness in their validity, there is an increased chance that data-driven methods will succeed as big data grows in popularity and analytics become advanced. To supplement changes, the theoretical foundations of data-driven science with regard to various methodologies and principles should be assessed and discussed to offer a new framework that accommodates the study details.<\/p>\n
The circumstances in big data and analytics are complicated due to the diversity of their theoretical foundations, with both being considered likely leads to the creation of other paradigms. Big data seems likely to improve the data suites available for assessments and allows for new techniques and approaches. This is due to theoretical positions and the likelihood that big data will be created to answer specific questions, thus increasing the need for more focused studies. Big data provides various opportunities for data-driven organizations to access vast amounts of data. It is a challenge since it results in a deficit in human skills required to assess and disseminate the data, thus developing an approach that allows for post-positivist types of computational disciplines.<\/p>\n
A probable way forward is an epistemology that is inspired by various statistics in the models, and approaches are utilized within a structure that is flexible and recognizes the positionality and situatedness of the topics being analyzed instead of rejecting the approaches. This type of epistemology has the ability to recognize and account for the utilization of abduction and developing a responsive data-driven organization. As the discussion shows, there is an increasing need for reflection on the epistemological effects of data analytics and big data.<\/p>\n
Research Design Structured Literature Review<\/p>\n
The research did a review of the literature on data management to discuss the first research question and get the key factors. To guarantee the requirements of the relevant study rigidity, brevity, clarity, breadth, and constituency (Levy & Ellis 2006), the assessment initiates the focus on study findings and theories to outline key issues. It seeks an unbiased description of extended coverage of the literature materials for scholars and professionals (vom Brocke et al., 2009). The four research databases chosen were IEEEXplore, AISeL, ACM, and Scopus since they have the appropriate Information System study articles and proceedings. Some of the keywords used include data-powered companies, data-driven organization, data-powered enterprise, data-driven enterprise (Klotzer & Pflaum, 2015). Various spelling criteria were also used, like distinguishing American and British English and spelling with and without hyphens. While there is a minimal chance reviewers can look at the information in this literature, it would be unethical to provide ideas by giving credit to the content owner(s). The concept of informed consent in relation to studies is a matter considered in the literature to design information gathering. The information obtained from other resources is used to carry out research on data and organizations.<\/p>\n
The suggested theory was called DECAS, or the concept enveloping the Decision-making procedure, and it consisted of Decision, dEcision maker, dAta, and analyticS. This is a quantitative concept that seeks to supplement previous ideas concerning the decision-making theories. In the literature, there are studies that have been identified and the various assumptions in the field through a comprehensive study assessment. The assumptions are analyzed based on the described data-driven choices scenarios, in which their disadvantages have been clearly stated. In this segment, the research develops its claims that are proven using previous research and looked at when considering the audience to create a contemporary decision concept. DECAS can be categorized as an analysis kind of theory (Kar &\u00a0Dwivedi, 2020). This theory states what is but fails to stretch past description and assessment and fails to explain or specify casual correlations or formulate forecasts. Additionally, DECAS seeks to assert the components inside a data-instigated decision-making domain and fails to stretch beyond describing the phenomena of concern and assessing the correlations between the constructs and the boundaries and scope in which the observations and correlations hold. Thus there are no casual foresight or correlations made or specified. The proposals focus on concepts and study findings in the domain of data-driven organizations to identify the main issues.<\/p>\n
Additionally, there are four key components of a theory, a way of illustration through which the concept is physically described in some way, the constructs that describe the interesting concept in theory, descriptions that indicate the correlation among the constructs, and the extent stipulated by the level of generality of the statements of boundaries and relationships indicating the boundaries of generalization. Also, Toulmin’s argumentation design was selected as an effective method in representing the elements of DECAS to frame the theory and show the constructs, scope, and statements of correlations through the model’s elements. The model has been considered suitable for application since it indicates a methodology for showing the alternative assumptions and analyzing them in relation to the research.<\/p>\n
There are six elements in Toulmin’s model that analyze the contentious framework for theories, recommendations, and propositions. The first three factors are necessary for any discussion, and they include the claim which stipulates the purpose of the discussion, the grounds also referred to as the basis for the discussion, facts, or supporting evidence, and the warrant which directly or indirectly supports the bases and correlates it to the claim. The other factors that can be added as relevant though not essential include the backing, which is the proof depended upon to establish and back up the relevance and reliability of the warrant, the qualifiers since they reduce the argument’s strength, and the conditions that would make the warrant seem invalid, and lastly the rebuttal which acknowledged those that may make a claim invalid.<\/p>\n
The components of DECAS include the representation of the theory using Toulmin’s argumentation model, the identification of the constructs and the claims being italicized, the warrants, grounds, and backings are utilized to define the statements and theory of the links between the constructs and the rebuttal and qualify applied in specifying the theory’s scope. Performance is vital in the era of Big Data, and the precise and rapid identification of data calls for better query systems and search engines (Aji et al., 2013). An approach that offers effective spatial query processing instead of various concurrent user queries and big spatial data is proposed. The method organizes spatial data while considering geographical closeness to accomplish increased input\/output throughput, then develops a double-tier dispensed spatial index for trimming the search space and applies an indexing + MapReduce architecture to improve the spatial query.<\/p>\n
The spatiotemporal big data has its challenges when looking at the computation and representation of time geographic relations and entities. There are proposed spatiotemporal models that apply a compressed linear reference method to change time geographic entities from a 3D road network to a 2Dcompressed linear reference space. This allows network time geographic entities to be managed and kept in a spatial database, and effective operations and index frameworks can be applied to carry out spatiotemporal queries and operations for these entities. While progress has been made, there needs to be additional research to hasten the big data access and recovery under different file systems.<\/p>\n
Sampling Procedures and or\/ Data Collection Sources <\/p>\n
The first set of 80 materials were located by looking at the keywords in the titles. Various studies were chosen that included data-driven innovation, data science, big data, data analytics, data-driven culture, and data-driven business models. The exclusion criteria involved selecting research materials that fell any of the following; the research left out studies outside the internet, those that fail to indicate any type of discussion or finding, it lacks any of the identified keywords in the introduction, it is not written in English, and lastly, if the study was a copy of another one. This left about 24 research materials that met all the criteria. The studies were used to develop a conceptual matrix that adheres to the rules written by Webster and Watson (2002). The concept matrix allowed the identification of the following factors; digital transformation, data-driven business models, data analytics, data science, and data-driven innovation.<\/p>\n
Data science enables the collection of data with granularity and high hope. Due to technological advances, data collection approaches are usually limited by the researcher’s imagination and not technological limitations. In fact, one of the difficulties is to think about how to create access to specified data for many observations. Big data gathering approaches that assist in overcoming challenges are web scraping, sensors, communications monitoring, and web traffic. The use of sensors allows one to constantly collect huge amounts of specific data. An instance is when organizations ask those within their workforce to wear wristbands that track activity. Data can be compiled on their progress and health while also collecting information on their proximity. Of importance is that the data-gathering approach is not obtrusive and offers information on employees in the work environment.<\/p>\n
Also, the methods enable the gathering of data over a long duration. Contrarily, optional approaches for getting data like diaries may impact behavior, thus making it hard to encourage employees to maintain a diary for a longer period (Aaker et al., 2013). Sensors can be effective in assessing the work environment. Also, there are companies that allow clients to assess the movement of the goods and services they order and the environment the goods or services are exposed to. Of course, when a company installs sensors to collect information on humans, it is essential to remain within the scope of the relevant research ethics and privacy laws. Even though this may seem normal, the complexities of gathering data together with problems like data ownership are complicated and thus require the understanding of guidelines used in study processes.<\/p>\n
Web scrapping enables the automated collection of vast amounts of data from the internet. Web scrapping initiatives are readily available, and they are free to use, like in the case of plugins for browsers. Also, one may utilize packages that are accessible in programming languages, and platforms like Twitter provide application program interfaces that streamline and ease one’s access to the avenue. Web scraping can be applied in gathering numeric information like prices, audio, textual, and visual data, or information on social media’s structure. The textual information that can be scraped includes social media content created by people and organizations, articles, and reviews. It is essential to know that before scraping data from the internet, it is essential to be careful when looking at the terms of use if they are available.<\/p>\n
Not all web pages agree to data scrapping since there are lawsuits that exist against organizations that participate in such endeavors (Reuters, 2013). Most of the cases touch on the commercial use of the scraped data. When doubts arise on the legality of scraping an organization’s data from the internet for academic purposes, it is appropriate to contact the website to get consent. Organizations monitor metrics like the number of those who have visited their web pages and the most visited content in a certain duration (Wiesel et al., 2011). Also, information on individual visitors can be accessed. Studies can get information on visitors, including the content that enabled them to reach the organization’s web page, the content looked at, and the order they were accessed. More data include the last content seen before leaving the platform, the duration users took accessing each page, the day and time they accessed the website, and whether there have been previous engagements between the user and the website. Data on the visitor enable the detailed assessment of how visitors like suppliers and investors operate on the company’s website, thus helping get the information requirements, their level of engagement with the organization, and how they intend to transact with the organization (Sismeiro & Bucklin, 2004). Apart from assessing traffic on the website, organizations can monitor outgoing traffic, that is, the web pages employees visit. Such information offer data on the external internet resources applied to employees and the duration spent on the internet for various purposes. Such information can be important in research when assessing employee well-being or product innovation.<\/p>\n
The information can also shift focus to employee engagement, attention allocation, and voluntary turnover. Information on communications between those within the workforce can be collected. An instance is the monitoring of email traffic and phone calls. Sources of this type of data assist in showing various internal networks and assess the flow of communication between branches, organizational departments, and various units at different hierarchy levels. Assessing internal and external phone, web, and email movement are approaches that are considered non-obtrusive, meaning users are unaware that the traffic created is under scrutiny thus it is vital to always adhere to laws on privacy and research ethics to safeguard consumers, employees, and other parties.<\/p>\n
The methods discussed enable the large-scale constant collection of data. Also, the features make it easy to execute various studies. In a randomized investigation, a randomly picked group of subjects like employees or consumers is exposed to a separate policy that the group tasked with controlling the study. If the attributes of the subjects are constantly tracked, the impact of the policy can be assessed (Lambrecht & Tucker, 2013). Experimental setups can be used, like random selection and the quasi-experimental method (Aaker et al., 2013). This subjects the agents to policy changes and how temporal variations can be used to analyze the impact of the changes. When doing so, it is essential not to consider alternative explanations.<\/p>\n
Statistical Tests<\/p>\n
Experiments are rare, and this could be attributed to the political and financial hurdles that they present. This is why most of the research is limited to a statistical assessment of observational data, which means information gathered without the use of a real experiment. This is referred to as a quasi-experiment since it looks at the organizational changes experienced as a result of transforming into a data-driven entity to estimate the effectiveness of data. Basically, the power of the findings from the research is limited. The fact that it is easy to collect and assess experimental data makes it possible to create situations that allow for the reexamination of the research question. The research did a meta-analysis on the keywords.<\/p>\n
The research used a diamond structure that is based on the previous study done by Leavitt (1965). The structure has a distinguished and long provenance in assessing DDOs and socio-technical systems (Vidgen et al., 2017). The model is used to assess the difficulties and offer an overview of data-driven organizations. The business insight was previously known as value creation since there are minimal guarantees that the research would result in increased value. For the purpose of the study, business insight and conceptual requirements were put in the same category since developing business understanding is an objective of any data-driven organization, while conceptual requirements have not been looked at by previous research.<\/p>\n
Summary<\/p>\n
The study tries to determine key factors of DDOs and conceptual needs. As shown in the study, a literature review is done. To answer the first research question, a joint structure indicating the factors of data-driven organizations and their distinct attributes. Conceptual needs are derived from literature and based on an organizational design concept to determine the answer to the second research question. The research analyzes and maintains a research scope that is relevant and has various study implications. The research suggests an avenue to develop an understanding of a data-driven organization and the factors it is made of. Each factor indicates an opportunity for more study projects that are valid, contextualized, and correlated based on this research’s structure.<\/p>\n
This is a problem through the abstraction of information and a contribution through its harmonizing and general scope. The findings can guide leaders in recognizing various action points in creating data-driven organizations. The work can be considered a checklist for important issues in the transformation procedures. This is done by focusing on managerial roles that focus on data management like the Chief Data Officer to assist in the transformation procedure more effectively. Consequently, there is a detailed value showing what constitutes data-driven organizations and providing an explanation on the areas that should be acted upon immediately. While looking for a complete sample, the research may have missed articles in the literature assessment. Also, the results are solely dependent on information derived from literature materials. Stretching the methodological range, for instance, by incorporating experimental examinations into data-driven organizations through questionnaires, interviews, or case studies in companies that have experienced the transition is a hindrance and reason for future studies. By gathering data from companies undergoing a transformation method, the key elements could be utilized to indicate the flow between critical factors that require many considerations as consequences and antecedents.<\/p>\n
In the research, we are shown the significance of data-instigated decision-making and why there needs to be more research in the field. There is a need to identify the benefits while mitigating the implications and challenges. Though the research tries to explain and solve a combination of factors and their challenges, there was a lack of evidence on the concepts that could encourage data-driven decision-making, thus making it a concept that is unique from other decision-making theories. The research seeks to engage in such studies by creating a new concept, DECAS that relies on various factors of decision theory and combines the analytics and data factors that characterize current data-driven environments. Also, the need for this type of concept, the concept, and the reasoning behind it, and its utilization are shown. The first claim is that the foundations of a data-driven decision include the decision-making procedure, the decision, the decision-maker, analytics, and big data. Another claim suggests the idea of collaborative rationality since, unlike the decision theory, which has limited rationality, it allows for the integration of machines and humans, thus optimizing the decision-making procedures (Kalantari, 2010). The final claim states that integrating these factors and selecting analytics and data can result in improved decisions.<\/p>\n
The significance of the study is the indication of a modern-day concept that can support various factors in data-instigated decision-making since known theories are not enough. The theory can be a basis for further studies and developments, especially in segments like machine learning, big data analytics, and metahuman systems. Also, the research is beneficial to the community and companies by indicating the factors required to promote data-instigated decision-making and how they can result in informed and quality decisions that would have been unachievable. This can add value to work done by those who are involved in data-driven decisions. Also, this is a simulated view of what can be accomplished by incorporating the factors of data-driven decision-making. This requires a lot of effort to create a meaningful combination between the factors. The amount of integration between machines and humans, the right selection of data, the proper application of analytics tools and approaches, and combining all these together is important, thus the need to research some to validate the claims further. Also, the conditions that result in accepting or rejecting data-driven decisions should be assessed.<\/p>\n
Besides, the significance of fields like traceability, reliability, governance, and accountability, which always elicit questions during decision-making, are becoming more relevant in data-driven decision-making. There are various topics that relate to the correlation between machines and human decision-makers. With the universal application of AI systems for encouraging human decision-making, the limited trust AI systems are subjected to should be addressed. This involves locating studies that make AI decisions genuine by offering evidence and the extent the evidence helps create trust (Schmidt et al., 2020). Also, with the creation of metahuman systems, there is a new change in studies towards machine learning and human collaboration and the variations they indicate (Lyytinen et al., 2020). Since future decision-making should be based on such collaborations, there should be careful studies.<\/p>\n
Future studies should concentrate on questions on whether the integration depends on the type of issue if the correlation between humans and machines is defined by specific decisions like tactical or strategic decisions, how decisions should be explained to both entities, who is to blame for errors made, and the types of errors or challenges that arise. How can decision-makers increase their efficacy in analytics and big data, focusing on concepts like uncertainty, accuracy, and errors, and lastly, how to create data-instigated decision-making that maintains a human presence as a semi-supervisor instead of being supervised by the machine. All these should be studied before the concept of data-driven organizations reaches its full potential.<\/p>\n
In the literature, we reviewed the use of data science in management. Scholars have opportunities to create improved ways of answering the current theories and new questions by adopting the data granularity and scope provided by big data. The starter kit describes the key issues affecting the gathering, storage, processing, and assessment of data, indicating a shift from normal approaches and paradigms. The field is undergoing rapid evolution in business operations and computing innovation. Thus the literature offers the basic factors needed to assess the data science methods. A few years back, the development of databases like UCINET and STATA changed management studies by allowing scholars to change to complex models that apply a vast amount of archival data.<\/p>\n
The arrival of data science is seen as the next stage in this development since it provides possibilities for refining existing concepts and improving the precision of existing empirical findings, the search for new research domains, and increasing kinds of research questions, resulting in an improved assessment that provides new insight on the methods that result in the observed impacts. Despite the increasing application of data science in management, marketing, and behavior research, their full potential is yet to be achieved, and the applications will require a lot of effort to create, refine, and optimize. <\/p>\n
Index<\/p>\n
Accountability, Pg. 14, 22, 105.<\/p>\n
Algorithm, Pg. 15, 29, 30, 32, 34, 43, 49, 51, 80.<\/p>\n
Artificial Intelligence, Pg. 59, 80.<\/p>\n
Automated decision-making Pg. 13, 90.<\/p>\n
Big data analytics Pg. 11, 33, 45, 46, 60, 67, 71, 73, 104. <\/p>\n
Business analytics Pg. 67.<\/p>\n
Communication Pg. 19, 21, 26, 28, 42, 47, 60, 80, 99,<\/p>\n
Computer Science Pg. 56, 57.<\/p>\n
Data-driven organizations Pg. 9, 10, 16, 19, 27, 37, 39. 42, 81, 82, 95, 102.<\/p>\n
Data Management Pg. 9, 19, 32, 42, 87, 90, 93, 104.<\/p>\n
Decision Theory Pg. 11, 15, 17, 24, 27, 34, 38, 42, 91, 104. <\/p>\n
Epistemological Pg. 17, 38, 90, 95.<\/p>\n
Information Systems Pg. 17, 68, 93, 96.<\/p>\n
Interactions Pg. 15, 21, 60, 71, 80.<\/p>\n
Phenomenon Pg. 80, 93<\/p>\n
Organizational performance Pg. 90.<\/p>\n
Social Media Pg. 48, 60, 64, 78, 100.<\/p>\n
Supply Chain Pg. 20, 60 <\/p>\n
Appendix A <\/p>\n
Tables<\/p>\n
Appendix B<\/p>\n
Figures<\/p>\n
Appendix C<\/p>\n
Consent Forms<\/p>\n
Appendix D<\/p>\n
IRB Approval<\/p>\n
References<\/p>\n
Agbozo, E. (2018). The role of data-driven e-government in realizing the sustainable development goals in developing economies. Journal of Information Systems & Operations Management, 12(1), 70-77.<\/p>\n
Agarwal, R., & Dhar, V. (2014). Big data, data science, and analytics: The opportunity and challenge for IS research.<\/p>\n
AlgorithmWatch. (2019).\u00a0Taking stock of automated decision-making in the EU. In\u00a0M.\u00a0Spielkamp\u00a0(Ed.),\u00a0Automating Society.\u00a0[Google Scholar]<\/p>\n
Amankwah-Amoah,\u00a0J., &\u00a0Adomako,\u00a0S.\u00a0(2019).\u00a0Big data analytics and business failures in data- rich environments: An organizing framework.\u00a0Computers in Industry,\u00a0105(2019),\u00a0204\u2013 212.\u00a0https:\/\/doi.org\/10.1016\/j.compind.2018.12.015\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Anderson, C. (2015). Creating a data-driven organization: Practical advice from the trenches. ” O’Reilly Media, Inc.”.<\/p>\n
Bartkus,\u00a0V.O.,\u00a0Mannor,\u00a0M.J.,\u00a0Campbell,\u00a0J.T., &\u00a0Crossland,\u00a0C.\u00a0(2018).\u00a0Fast and rigorous: Configurational determinants of strategic decision-making balance. In\u00a0Academy of Management Proceedings\u00a0(Vol. 2008, No. 1).\u00a0Academy of Management.\u00a0https:\/\/doi.org\/10.5465\/AMBPP.2018.264\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Bean,\u00a0R., &\u00a0Davenport,\u00a0T.H.\u00a0(2019).\u00a0Companies are failing in their efforts to become data- driven.\u00a0Harvard Business Review.\u00a0Harvard Business Publishing.\u00a0[Google Scholar]<\/p>\n
Bell,\u00a0D.E.,\u00a0Raiffa,\u00a0H., &\u00a0Tversky,\u00a0A.\u00a0(1988).\u00a0Decision making: Descriptive, normative, and prescriptive interactions. Chapter 2.\u00a0Cambridge University Press.\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Boyd,\u00a0D., &\u00a0Crawford,\u00a0K.\u00a0(2012).\u00a0Critical questions for big data: Provocations for a cultural, technological, and scholarly phenomenon.\u00a0Information, Communication & Society, 15(5),\u00a0662\u2013679.\u00a0https:\/\/doi.org\/10.1080\/1369118X.2012.678878\u00a0[Taylor & Francis Online],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Buchanan,\u00a0L., &\u00a0O’Connell,\u00a0A.\u00a0(2006).\u00a0A brief history of decision-making.\u00a0Harvard Business Review, 84(1).\u00a0[PubMed],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Burton,\u00a0J.W.,\u00a0Stein,\u00a0M.K., &\u00a0Jensen,\u00a0T.B.\u00a0(2020).\u00a0A systematic review of algorithm aversion in augmented decision making.\u00a0Journal of Behavioral Decision Making, 33(2),\u00a0220\u2013 239.\u00a0https:\/\/doi.org\/10.1002\/bdm.2155\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Cao,\u00a0G., &\u00a0Duan,\u00a0Y.\u00a0(2015).\u00a0The affordances of business analytics for strategic decision-making and their impact on organisational performance. In\u00a0Proceedings of the 19th Pacific Asia Conference on Information Systems (PACIS 2015),\u00a0Singapore.\u00a0[Google Scholar]<\/p>\n
Davenport, T. H., Barth, P., & Bean, R. (2012).\u00a0How’big data’is different. MIT Sloan Management Review.<\/p>\n
Diakopoulos,\u00a0N.\u00a0(2016).\u00a0Accountability in algorithmic decision making.\u00a0Communications of the ACM, 59(2),\u00a056\u201362.\u00a0https:\/\/doi.org\/10.1145\/2844110\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Doyle,\u00a0J., &\u00a0Thomason,\u00a0R.H.\u00a0(1999).\u00a0Background to qualitative decision theory.\u00a0AI Magazine, 20(22),\u00a055\u201368.\u00a0https:\/\/doi.org\/10.1609\/aimag.v20i2.1456\u00a0[Google Scholar]<\/p>\n
Dremel,\u00a0C.,\u00a0Herterich,\u00a0M.,\u00a0Wulf,\u00a0J.,\u00a0Waizmann,\u00a0J.-C., &\u00a0Brenner,\u00a0W.\u00a0(2017).\u00a0How AUDI AG established big data analytics in its digital transformation.\u00a0MIS Quarterly Executive, 16(2), 81-100.\u00a0https:\/\/aisel.aisnet.org\/misqe\/vol16\/iss2\/3\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Drucker,\u00a0P.F.\u00a0(1967).\u00a0The effective decision.\u00a0Harvard Business Review, 45(1),\u00a092\u201398.\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Duan,\u00a0Y.,\u00a0Edwards,\u00a0J.S., &\u00a0Dwivedi,\u00a0Y.K.\u00a0(2019).\u00a0Artificial intelligence for decision making in the era of big data\u2013evolution, challenges and research agenda.\u00a0International Journal of Information Management,\u00a048(2019),\u00a06371.\u00a0https:\/\/doi.org\/10.1016\/j.ijinfomgt.2019.01.021\u00a0[Crossref], \u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Dubey, R., Gunasekaran, A., and Childe, S. J. 2019. \u201cBig data analytics capability in supply chain agility,\u201d Management Decision (8), pp. 2092-2112 (doi: 10.1108\/MD-01-2018- 0119).<\/p>\n
Elgendy,\u00a0N., &\u00a0Elragal,\u00a0A.\u00a0(2014).\u00a0Big data analytics: A literature review paper. In\u00a0P.\u00a0Perner\u00a0(Ed.),\u00a0Advances in data mining, (pp.\u00a0214\u2013227, Applications and Theoretical Aspects,\u00a0Lecture Notes in Computer Science, 8557).\u00a0Springer, Cham.\u00a0https:\/\/doi.org\/10.1007\/978-3-319-08976-8_16.\u00a0[Google Scholar]<\/p>\n
Elgendy,\u00a0N., &\u00a0Elragal,\u00a0A.\u00a0(2016).\u00a0Big data analytics in support of the decision-making process.\u00a0Procedia Computer Science,\u00a0100(2016),\u00a01071\u2013 1084.\u00a0https:\/\/doi.org\/10.1016\/j.procs.2016.09.251\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Elragal,\u00a0A., &\u00a0Klischewski,\u00a0R.\u00a0(2017).\u00a0Theory-driven or process-driven prediction? Epistemological challenges of big data analytics.\u00a0Journal of Big Data, 4(1),\u00a01\u2013 20.\u00a0https:\/\/doi.org\/10.1186\/s40537-017-0079-2\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
European Commission 2020. A European strategy for data, Brussels.<\/p>\n
Falletta, S. 2014. \u201cOrganizational Diagnostic Models – A Review and Synthesis,\u201d Organizational Intelligence Institute. <\/p>\n
Frantz,\u00a0R.\u00a0(2003).\u00a0Herbert Simon. Artificial intelligence as a framework for understanding intuition.\u00a0Journal of Economic Psychology, 24(2),\u00a0265\u2013 277.\u00a0https:\/\/doi.org\/10.1016\/S0167-4870(02)00207-6\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Frisk,\u00a0J.E., &\u00a0Bannister,\u00a0F.\u00a0(2017).\u00a0Improving the use of analytics and big data by changing the decision-making culture: A design approach.\u00a0Management Decision, 55(10),\u00a02074\u2013 2088.\u00a0https:\/\/doi.org\/10.1108\/MD-07-2016-0460\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Galbraith, J. R. 2016. \u201cTHE STAR MODEL\u2122,\u201d <\/p>\n
Gandomi, A., & Haider, M. (2015). Beyond the hype: Big data concepts, methods, and analytics.\u00a0International journal of information management,\u00a035(2), 137-144.<\/p>\n
Gigerenzer,\u00a0G., &\u00a0Gaissmaier,\u00a0W.\u00a0(2011).\u00a0Heuristic decision making.\u00a0Annual Review of Psychology, 62(1),\u00a0451\u2013482.\u00a0https:\/\/doi.org\/10.1146\/annurev-psych-120709- 145346\u00a0[Crossref],\u00a0[PubMed],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Gigerenzer,\u00a0G., &\u00a0Gaissmaier,\u00a0W.\u00a02015.\u00a0Decision making: Nonrational theories.\u00a0International encyclopedia of the social & behavioral sciences\u00a0(2nd. Vol. 5, pp.\u00a0911\u2013916). Elsevier.\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Govindan, K., Cheng, T. E., Mishra, N., & Shukla, N. (2018). Big data analytics and application for logistics and supply chain management.<\/p>\n
Grabo\u015b,\u00a0R.\u00a0(2004).\u00a0Qualitative model of decision making. in\u00a0International Conference on Artificial Intelligence: Methodology, Systems, and Applications,\u00a0Berlin, Heidelberg:\u00a0Springer,\u00a0480\u2013489\u00a0(https:\/\/doi.org\/10.1007\/978-3-540-30106- 6_49).\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Grant, R. M. 1996. \u201cToward a knowledge\u2010based theory of the firm,\u201d Strategic Management Journal. <\/p>\n
Gregor,\u00a0S.\u00a0(2006).\u00a0The nature of theory in information systems.\u00a0MIS Quarterly, 30(3),\u00a0611\u2013 642.\u00a0https:\/\/doi.org\/10.2307\/25148742\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Grover,\u00a0P., &\u00a0Kar,\u00a0A.K.\u00a0(2017).\u00a0Big data analytics: A review on theoretical contributions and tools used in literature.\u00a0Global Journal of Flexible Systems Management, 18(3),\u00a0203\u2013 229.\u00a0https:\/\/doi.org\/10.1007\/s40171-017-0159-3\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Grover,\u00a0P.,\u00a0Kar,\u00a0A.K., &\u00a0Dwivedi,\u00a0Y.K.\u00a0(2020).\u00a0Understanding artificial intelligence adoption in operations management: Insights from the review of academic literature and social media discussions.\u00a0Annals of Operations Research, 2020,\u00a01\u201337.\u00a0https:\/\/doi.org\/10.1007\/s10479- 020-03683-9\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Grover,\u00a0V.,\u00a0Lindberg,\u00a0A.,\u00a0Benbasat,\u00a0I., &\u00a0Lyytinen,\u00a0K.\u00a0(2020).\u00a0The perils and promises of big data research in information systems.\u00a0Journal of the Association for Information Systems, 21(2),\u00a0268\u2013291.\u00a0https:\/\/doi.org\/10.17705\/1jais.00601\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Grover,\u00a0V., &\u00a0Lyytinen,\u00a0K.\u00a0(2015).\u00a0New state of play in information systems research: The push to the edges.\u00a0MIS Quarterly, 39(2),\u00a0271\u2013 296.\u00a0https:\/\/doi.org\/10.25300\/MISQ\/2015\/39.2.01\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Gupta,\u00a0S.,\u00a0Kar,\u00a0A.K.,\u00a0Baabdullah,\u00a0A., &\u00a0Al-Khowaiter,\u00a0W.A.\u00a0(2018).\u00a0Big data with cognitive computing: A review for the future.\u00a0International Journal of Information Management,\u00a042(2018),\u00a078\u2013<\/p>\n
Guggenberger, T., M\u00f6ller, F., Boualouch, K., and Otto, B. 2020. \u201cTowards a Unifying Understanding of Digital Business Models,\u201d 24th Pacific Asia Conference on Information Systems, pp. 1-14.<\/p>\n
89.\u00a0https:\/\/doi.org\/10.1016\/j.ijinfomgt.2018.06.005\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Habermas,\u00a0J.\u00a0(1984).\u00a0The theory of communicative action, vol. 1, reason and the rationalization of society.\u00a0Heinemann.\u00a0[Google Scholar]<\/p>\n
Hansson,\u00a0S.O.\u00a0(1994).\u00a0Decision theory. A brief introduction. In\u00a0Department of philosophy and the history of technology. Stockholm:\u00a0Royal Institute of Technology.\u00a0[Google Scholar]<\/p>\n
Hansson,\u00a0S.O.\u00a0(2011).\u00a0Decision theory: An overview. In\u00a0M.\u00a0Lovric\u00a0(Ed.),\u00a0International encyclopedia of statistical science\u00a0(pp.\u00a0349\u2013355).\u00a0Springer.\u00a0https:\/\/doi.org\/10.1007\/978- 3-642-04898-2_22\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Hassan,\u00a0N.R., &\u00a0Mingers,\u00a0J.\u00a0(2018).\u00a0Reinterpreting the kuhnian paradigm in information systems.\u00a0Journal of the Association for Information Systems, 19(7),\u00a0568\u2013 599.\u00a0https:\/\/doi.org\/10.17705\/1jais.00502\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Hartmann, P., Zaki, M., and Feldmann, Niels, Neely, A. 2016. \u201cCapturing value from big data: A taxonomy of data-driven business models used by start-up firms,\u201d International journal of operations & production management, pp. 1-19. <\/p>\n
Henderson, D., Earley, S., Sebastian-Coleman, L., Sykora, E., and Smith, E. 2017. DAMA- DMBOK: Data management body of knowledge, Basking Ridge, New Jersey: Technics Publications. <\/p>\n
Hemerly, J. (2013). Public policy considerations for data-driven innovation. Computer, 46(6), 25-31.<\/p>\n
Hernaus, T., Aleksic, A., and Klindzic, M. 2013. \u201cOrganizing for Competitiveness \u2013 Structural and Process Characteristics of Organizational Design,\u201d Contemporary Economics (7:4), pp. 25-40 (doi: 10.5709\/ce.1897-9254.122). <\/p>\n
Ho,\u00a0A.T.\u00a0(2017).\u00a0Big data and evidence-driven decision-making: Analyzing the practices of large and mid-sized US cities. In\u00a0Proceedings of the 50th Hawaii International Conference on System Sciences\u00a0(pp.\u00a02794\u20132903).\u00a0Hawaii.\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Innes,\u00a0J.E., &\u00a0Booher,\u00a0D.E.\u00a0(2018).\u00a0Planning with complexity: An introduction to collaborative rationality for public policy.\u00a0Routledge.\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Intezari,\u00a0A., &\u00a0Gressel,\u00a0S.\u00a0(2017).\u00a0Information and reformation in KM systems: Big data and strategic decision-making.\u00a0Journal of Knowledge Management, 21(1),\u00a071\u2013 91.\u00a0https:\/\/doi.org\/10.1108\/JKM-07-2015-0293\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Ioannidis,\u00a0J.P.,\u00a0Cripps,\u00a0S., &\u00a0Tanner,\u00a0M.A.\u00a0(2020).\u00a0Forecasting for COVID-19 has failed.\u00a0International Journal of Forecasting.\u00a0https:\/\/doi.org\/10.1016\/j.ijforecast.2020.08.004\u00a0[Crossref],\u00a0[PubMed],\u00a0[Goo gle Scholar]<\/p>\n
Janssen,\u00a0M.,\u00a0Van Der Voort,\u00a0H., &\u00a0Wahyudi,\u00a0A.\u00a0(2017).\u00a0Factors influencing big data decision- making quality.\u00a0Journal of Business Research,\u00a070(2017),\u00a0338\u2013 345.\u00a0https:\/\/doi.org\/10.1016\/j.jbusres.2016.08.007\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Jha,\u00a0A.K.,\u00a0Agi,\u00a0M.A., &\u00a0Ngai,\u00a0E.W.\u00a0(2020).\u00a0A note on big data analytics capability development in supply chain.\u00a0Decision Support Systems,\u00a0138(2020),\u00a0113382.\u00a0https:\/\/doi.org\/10.1016\/j.dss.2020.113382\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Jurgen, R. K. 1983. \u201cData-driven automation,\u201d IEEE Spectrum (20:5), pp. 33-35 (doi: 10.1109\/MSPEC.1983.6369900).<\/p>\n
Kates, A., and Galbraith, J. R. 2007. Designing your organization: Using the star model to solve 5 critical design challenges, San Francisco: Jossey-Bass. <\/p>\n
Kahneman,\u00a0D.\u00a0(2003).\u00a0Maps of bounded rationality: Psychology for behavioral economics.\u00a0American Economic Review, 93(5),\u00a01449\u2013 1475.\u00a0https:\/\/doi.org\/10.1257\/000282803322655392\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Kalantari,\u00a0B.\u00a0(2010).\u00a0Herbert a. Simon on making decisions: Enduring insights and bounded rationality.\u00a0Journal of Management History, 16(4),\u00a0509\u2013 520.\u00a0https:\/\/doi.org\/10.1108\/17511341011073988\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Kamioka,\u00a0T., &\u00a0Tapanainen,\u00a0T.\u00a0(2014).\u00a0Organizational use of big data and competitive advantage-exploration of antecedents.\u00a0Pacific Asia Conference on Information Systems (PACIS 2014). China.\u00a0[Google Scholar]<\/p>\n
Kar,\u00a0A.K., &\u00a0Dwivedi,\u00a0Y.K.\u00a0(2020).\u00a0Theory building with big data-driven research\u2013moving away from the \u201cWhat\u201d towards the \u201cWhy\u201d.\u00a0International Journal of Information Management,\u00a054(2020),\u00a0102-205.\u00a0https:\/\/doi.org\/10.1016\/j.ijinfomgt.2020.102205\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Karbach,\u00a0J.\u00a0(1987).\u00a0Using Toulmin\u2019s model of argumentation.\u00a0Journal of Teaching Writing, 6(1),\u00a081\u201392.\u00a0[Google Scholar]<\/p>\n
Kearny, C., Gerber, A., and van der Merwe, A. 2016. \u201cData-driven enterprise architecture and the TOGAF ADM phases,\u201d in 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Budapest, Hungary, IEEE, pp. 4603-4608. <\/p>\n
Kim, M., Zimmermann, T., DeLine, R., & Begel, A. (2016, May). The emerging role of data scientists on software development teams. In 2016 IEEE\/ACM 38th International Conference on Software Engineering (ICSE) (pp. 96-107). IEEE.<\/p>\n
Klotzer, C., and Pflaum, A. 2015. \u201cCyber-physical systems as the technical foundation for problem solutions in manufacturing, logistics and supply chain management,\u201d in 2015 5th International Conference on the Internet of Things (IOT), Seoul, IEEE, pp. 12-19.<\/p>\n
Kotsiantis,\u00a0S.B.,\u00a0Zaharakis,\u00a0I.D., &\u00a0Pintelas,\u00a0P.E.\u00a0(2006).\u00a0Machine learning: A review of classification and combining techniques.\u00a0Artificial Intelligence Review, 26(3),\u00a0159\u2013 190.\u00a0https:\/\/doi.org\/10.1007\/s10462-007-9052-3\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Langley,\u00a0A.,\u00a0Mintzberg,\u00a0H.,\u00a0Pitcher,\u00a0P.,\u00a0Posada,\u00a0E., &\u00a0Saint-Macary,\u00a0J.\u00a0(1995).\u00a0Opening up decision making: The view from the black stool.\u00a0Organization Science, 6(3),\u00a0260\u2013 279.\u00a0https:\/\/doi.org\/10.1287\/orsc.6.3.260\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Leavitt, H. J. 1965. \u201cApplying organizational change in industry: Structual, technological and humanistic approaches,\u201d Handbook of organizations, pp. 1144-1170. <\/p>\n
Leavitt, H. J., and March, J. G. 1962. \u201cApplied Organizational Change in Industry: Structural, Technological and Humanistic Approaches,\u201d Modern Economy (5). <\/p>\n
Levitt, R. E., Thomsen, J., Christiansen, T. R., Kunz, J. C., Jin, Y., and Nass, C. 1999. \u201cSimulating Project Work Processes and Organizations: Toward a Micro-Contingency Theory of Organizational Design,\u201d Management Science (45:11), pp. 1479-1495 (doi: 10.1287\/mnsc.45.11.1479). <\/p>\n
Lyytinen,\u00a0K., &\u00a0Grover,\u00a0V.\u00a0(2017).\u00a0Management misinformation systems: A time to revisit?\u00a0Journal of the Association for Information Systems, 18(3),\u00a0206\u2013 230.\u00a0https:\/\/doi.org\/10.17705\/1jais.00453\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Lyytinen,\u00a0K.,\u00a0Nickerson,\u00a0J.V., &\u00a0King,\u00a0J.L.\u00a0(2020).\u00a0Metahuman systems= humans+ machines that learn.\u00a0Journal of Information Technology,\u00a01\u2013 19.\u00a0https:\/\/doi.org\/10.1177\/0268396220915917\u00a0[Google Scholar]<\/p>\n
Mandinach,\u00a0E.B.\u00a0(2012).\u00a0A perfect time for data use: Using data-driven decision making to inform practice.\u00a0Educational Psychologist, 47(2),\u00a071\u2013 85.\u00a0https:\/\/doi.org\/10.1080\/00461520.2012.667064\u00a0[Taylor & Francis Online],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Matheus, R., Janssen, M., & Maheshwari, D. (2018). Data science empowering the public: Data- driven dashboards for transparent and accountable decision-making in smart cities. Government Information Quarterly, 101284.<\/p>\n
Maalej, W., Nayebi, M., Johann, T., & Ruhe, G. (2015). Toward data-driven requirements engineering. IEEE Software, 33(1), 48-54.<\/p>\n
Manyika,\u00a0J.,\u00a0Chui,\u00a0M.,\u00a0Brown,\u00a0B.,\u00a0Bughin,\u00a0J.,\u00a0Dobbs,\u00a0R.,\u00a0Roxburgh,\u00a0C., &\u00a0Byers,\u00a0A.H.\u00a0(2011).\u00a0Big data: The next frontier for innovation, competition, and productivity.\u00a0McKinsey Global Institute Reports,\u00a01\u2013156. McKinsey Global Institute.\u00a0[Google Scholar]<\/p>\n
McAfee, A., Brynjolfsson, E., Davenport, T. H., Patil, D. J., & Barton, D. (2012). Big data: the management revolution.\u00a0Harvard business review,\u00a090(10), 60-68.<\/p>\n
Mezias,\u00a0J., &\u00a0Starbuck,\u00a0W.H.\u00a0(2009).\u00a0Decision making with inaccurate, unreliable data. In\u00a0The Oxford handbook of organizational decision making\u00a0(pp. 76-96).\u00a0Oxford University Press.\u00a0[Google Scholar]<\/p>\n
Mikalef,\u00a0P.,\u00a0Pappas,\u00a0I.O.,\u00a0Krogstie,\u00a0J., &\u00a0Giannakos,\u00a0M.\u00a0(2018).\u00a0Big data analytics capabilities: A systematic literature review and research agenda.\u00a0Information Systems and E-Business Management, 16(3),\u00a0547\u2013578.\u00a0https:\/\/doi.org\/10.1007\/s10257-017-0362- y\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Mintzberg,\u00a0H.\u00a0(1975).\u00a0The Manager\u2019s job: Folklore and fact.\u00a0Harvard Business Review,\u00a053(1990),\u00a049\u201361.\u00a0[Google Scholar]<\/p>\n
Mintzberg,\u00a0H.\u00a0(1989).\u00a0On management. In\u00a0Mintzberg on management: Inside our strange world of organizations.\u00a0Simon and Schuster.\u00a0[Google Scholar]<\/p>\n
Mintzberg,\u00a0H., &\u00a0Westley,\u00a0F.\u00a0(2001).\u00a0Decision making: It\u2019s not what you think.\u00a0MIT Sloan Management Review, 42(3),\u00a089\u201393.\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Nahapiet, J., and Ghoshal, S. 1998. \u201cSocial Capital, Intellectual Capital, and the Organizational Advantage,\u201d The Academy of Management Review (23), pp. 242-266. <\/p>\n
NewVantage Partners (2019).\u00a0Data and innovation. how big data and AI are accelerating business transformation.\u00a0Big Data and AI Executive Survey 2019. NewVantage Partners.\u00a0[Google Scholar]<\/p>\n
Niederman, F., and March, S. 2019. \u201cThe “Theoretical Lens” Concept: We All Know What it Means, but do We All Know the Same Thing?\u201d Commun. Assoc. Inf. Syst. (44:1), pp. 1- 33 (doi: 10.17705\/1CAIS.04401). <\/p>\n
Nutt,\u00a0P. C.\u00a0(2010). Building a decision making action theory.\u00a0Handbook of Decision Making\u00a0(pp.\u00a0155\u2013196).\u00a0Chichester:\u00a0John Wiley & Sons.\u00a0[Google Scholar]<\/p>\n
Oliveira, M., Lima, G., and L\u00f3scio, B. 2019. \u201cInvestigations into Data Ecosystems: a systematic mapping study,\u201d Knowledge and Information Systems (61:2), pp. 589-630 (doi: 10.1007\/s10115-018-1323-6). <\/p>\n
Olszak, C., and Zurada, J. 2019. \u201cBig Data-driven Value Creation for Organizations,\u201d Proceedings of the 52nd Hawaii International Conference on System Sciences, pp. 164- 173.<\/p>\n
Paschen,\u00a0J.,\u00a0Wilson,\u00a0M., &\u00a0Ferreira,\u00a0J.J.\u00a0(2020).\u00a0Collaborative intelligence: How human and artificial intelligence create value along the B2B sales funnel.\u00a0Business Horizons, 63(3),\u00a0403\u2013414.\u00a0https:\/\/doi.org\/10.1016\/j.bushor.2020.01.003\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Peterson,\u00a0M.\u00a0(2011).\u00a0Decision theory: An introduction. In\u00a0M.\u00a0Lovric M.\u00a0(Ed.),\u00a0International encyclopedia of statistical science\u00a0(pp.\u00a0349\u2013356).\u00a0Springer.\u00a0https:\/\/doi.org\/10.1007\/978- 3-642-04898-2_23\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Pomerol,\u00a0J.C., &\u00a0Adam,\u00a0F.\u00a0(2004).\u00a0Practical decision making \u2013 From the legacy of Herbert Simon to decision support systems. in\u00a0Proceedings of the Decision Support in an Uncertain and Complex World: The IFIP TC8\/WG8.3\u00a0International Conference,\u00a0647\u2013 657.\u00a0[Google Scholar]<\/p>\n
Polzonetti, A., and Sagratella, M. 2017. \u201cTowards a Data-Driven Enterprise: Effects on Information, Governance, Infrastructures and Security,\u201d IEEE Transactions on Engineering Management, pp. 1480-1484. <\/p>\n
Power,\u00a0D.J.\u00a0(2016).\u00a0Data science: Supporting decision-making.\u00a0Journal of Decision Systems, 25(4),\u00a0345\u2013356.\u00a0https:\/\/doi.org\/10.1080\/12460125.2016.1171610\u00a0[Taylor & Francis Online],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Power,\u00a0D.J.,\u00a0Cyphert,\u00a0D., &\u00a0Roth,\u00a0R.M.\u00a0(2019).\u00a0Analytics, bias, and evidence: The quest for rational decision making.\u00a0https:\/\/doi.org\/10.1080\/12460125.2019.1623534\u00a0[Taylor & Francis Online],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Power,\u00a0D.J.,\u00a0Heavin,\u00a0C., &\u00a0Keenan,\u00a0P.\u00a0(2019).\u00a0Decision systems redux.\u00a0Journal of Decision Systems, 28(1),\u00a01\u201318.\u00a0https:\/\/doi.org\/10.1080\/12460125.2019.1631683\u00a0[Taylor & Francis Online],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Provost,\u00a0F., &\u00a0Fawcett,\u00a0T.\u00a0(2013).\u00a0Data science and its relationship to big data and data-driven decision making.\u00a0Big Data, 1(1),\u00a051\u2013 59.\u00a0https:\/\/doi.org\/10.1089\/big.2013.1508\u00a0[Crossref],\u00a0[PubMed],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Ransbotham,\u00a0S.,\u00a0Khodabandeh,\u00a0S.,\u00a0Kiron,\u00a0D.,\u00a0Candelon,\u00a0F.,\u00a0Chu,\u00a0M., &\u00a0LaFountain,\u00a0B.\u00a0(2020),\u00a0Expanding AI\u2019s impact with organizational learning.\u00a0MIT Sloan Management Review and Boston Consulting Group.\u00a0[Google Scholar]<\/p>\n
Rathore,\u00a0A.K.,\u00a0Kar,\u00a0A.K., &\u00a0Ilavarasan,\u00a0P.V.\u00a0(2017).\u00a0Social media analytics: Literature review and directions for future research.\u00a0Decision Analysis, 14(4),\u00a0229\u2013 249.\u00a0https:\/\/doi.org\/10.1287\/deca.2017.0355\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Reinsel, D., Gantz, J., and Rydning, J. 2018. The Digitization of the World from Edge to Core, Framingham, MA: IDC.<\/p>\n
Saeed, B. B., and Wang, W. 2013. \u201cOrganisational diagnoses: a survey of the literature and proposition of a new diagnostic model,\u201d Int. J. Information Systems and Change Management, (6), pp. 222-238. <\/p>\n
Saggi,\u00a0M.K., &\u00a0Jain,\u00a0S.\u00a0(2018).\u00a0A survey towards an integration of big data analytics to big insights for value-creation.\u00a0Information Processing & Management, 54(5),\u00a0758\u2013 790.\u00a0https:\/\/doi.org\/10.1016\/j.ipm.2018.01.010\u00a0[Crossref],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Sandrin, E., Trentin, A., and Forza, C. 2014. \u201cOrganizing for Mass Customization: Literature Review and Research Agenda,\u201d International Journal of Industrial Engineering and Management (5), pp. 159-167. <\/p>\n
Shamim, S., Zeng, J., Shariq, S. M., and Khan, Z. 2019. \u201cRole of big data management in enhancing big data decision-making capability and quality among Chinese firms: A dynamic capabilities view,\u201d Information & Management (56:6) (doi: 10.1016\/j.im.2018.12.003). <\/p>\n
Shanks, G., and Bekmamedova, N. 2012. \u201cThe Impact of Strategy on Business Analytics Success,\u201d ACIS 2012 Proceedings. <\/p>\n
Schmidt,\u00a0P.,\u00a0Biessmann,\u00a0F., &\u00a0Teubner,\u00a0T.\u00a0(2020).\u00a0Transparency and trust in artificial intelligence systems.\u00a0Journal of Decision Systems, 29(4),\u00a0260\u2013265<\/p>\n
Sch\u00fcritz, R., Brand, E., Satzger, G., and Bischoffshausen, J. 2017. \u201cHow to cultivate analytics capabilities within an organization? Design and types of analytics competency centers,\u201d in ECIS 2017 Proceedings. What is a Data-Driven Organization? Twenty-Seventh Americas Conference on Information Systems, Montreal, 2021 10 <\/p>\n
Simon,\u00a0H.A.\u00a0(1959).\u00a0Theories of decision-making in economics and behavioral science.\u00a0The American Economic Review, 49(3),\u00a0253\u2013283.\u00a0https:\/\/doi.org\/10.1007\/978-1-349-00210- 8_1\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n
Simon,\u00a0H.A.\u00a0(1977).\u00a0Thinking by computers.\u00a0Models of Discovery. Boston Studies in the Philosophy of Science, 54,\u00a0268\u2013285.\u00a0https:\/\/doi.org\/10.1007\/978-94-010-9521- 1_15\u00a0[Google Scholar]<\/p>\n
Simon,\u00a0H.A.\u00a0(1997).\u00a0Models of bounded rationality: Empirically grounded economic reason vol. 3.\u00a0MIT Press.\u00a0[Crossref],\u00a0[Google Scholar]<\/p>\n
Storm, M., and Borgman, H. 2020. \u201cUnderstanding challenges and success factors in creating a data-driven culture,\u201d Proceedings of the 53rd Hawaii International Conference on System Sciences, pp. 5399-5408.<\/p>\n
Strand,\u00a0M., &\u00a0Syberfeldt,\u00a0A.\u00a0(2020).\u00a0Using external data in a BI solution to optimise waste management.\u00a0Journal of Decision Systems, 29(1),\u00a053\u2013 68.\u00a0https:\/\/doi.org\/10.1080\/12460125.2020.1732174\u00a0[Taylor & Francis Online],\u00a0[Web of Science \u00ae],\u00a0[Google Scholar]<\/p>\n","protected":false},"excerpt":{"rendered":"
Role Of Data Science In Making Data-Driven Decisions In App Development Companies 2 Role Of Data Science In Making Data-Driven Decisions In App Development Companies 2 THESIS PROPOSAL TEMPLATE Title: ROLE OF DATA SCIENCE IN MAKING DATA-DRIVEN DECISIONS IN APP DEVELOPMENT COMPANIES Prepared by Rajyalakshmi Kommineni University of the Cumberland\u2019s Ph.D. DATE:11\/26\/2021 [Cover Page] Title: […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[10],"class_list":["post-78123","post","type-post","status-publish","format-standard","hentry","category-research-paper-writing","tag-writing"],"_links":{"self":[{"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/posts\/78123","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/comments?post=78123"}],"version-history":[{"count":0,"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/posts\/78123\/revisions"}],"wp:attachment":[{"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/media?parent=78123"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/categories?post=78123"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/papersspot.com\/blog\/wp-json\/wp\/v2\/tags?post=78123"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}