Coupling Data Science and Design Science to Solve Real-World Business and Healthcare Challenges
Publisher
The University of Arizona.Rights
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.Embargo
Release after 06/08/2028Abstract
Big and heterogeneous data that have become increasingly available provide precious opportunities for addressing real-world challenges. Nevertheless, the 3Vs (volume, velocity and variety) of big data have created computational challenges. In addition, because big data are often not collected for the purpose of research, careful repurposing is needed in harnessing the power of data. Data science has provided powerful tools to address the 3Vs of big data, while design science offers a paradigm that guide the effective usage of tools and the repurpose of data. In this dissertation, three essays are included to demonstrate how data science and design science may be coupled to address real-world business and healthcare challenges. In the first essay, a theory-enhanced hierarchical neural network model with correction is proposed to provide fine-grained classification of social media users. The artifact is an important tool that helps the repurposing of social media data and is itself a demonstration of coupling data science tools (e.g., machine learning) with design science (e.g., theory-guided design). The second essay integrates and repurposes heterogenous data sources from the contact tracing process to evaluate multi-method contact tracing. It is another manifestation of using careful design with the help of analytical tools to obtain insights from data. The third essay investigates the use of theoretical lenses in startup success prediction. Specifically, the effect of social capital theory and knowledge spillover theory is experimented. It demonstrates how theories can guide the design choices and enhance predictive modeling that used to be data-centric. This dissertation has demonstrated how data science and design science can be integrated to address real-world business and healthcare challenges. It has great relevance to both the data science and designs science communities, and the artifacts and insights created also have great implications to researchers and practitioners in the field of social media analytics, contact tracing, and startup evaluating and investing.Type
Electronic Dissertationtext
Degree Name
Ph.D.Degree Level
doctoralDegree Program
Graduate CollegeManagement Information Systems