Theory-Informed Design and Evaluation of Web-Based Knowledge Management Systems

With the rapid increase in information availability, it is nearly impossible to stay current with advances in knowledge. The overarching question that my dissertation seeks to address is how to design Web-based knowledge management systems (KMS) to alleviate this problem by providing search and analysis support for individuals to access and understand the accumulated knowledge in an effective and efficient way. Specifically, designing and implementing consistent representation of heterogeneous data, enabling efficient and user-friendly search support, and supporting advanced user-interactive analysis and visualization are important issues to address. Certain Information Systems and cognitive theories can be leveraged to guide the design and evaluation of these systems. In addition, the evaluation results can further validate and extend theory. This helps to bridge the design-science paradigm and the behavioral-science paradigm.The research studies in my dissertation involve the development of Web-based KMS based on different types of data sources. Such systems aim to support decision making and can be important for researchers, practitioners, business investors, and policy makers. Chapter 2 develops a Web-based KMS, Arizona Literature Mapper, which allows users to gain comprehensive understanding of bioterrorism research. The data source of Arizona Literature Mapper is scientific publications. The system integrates techniques for content analysis, network analysis, and information visualization. Two user studies were conducted to evaluate the performance of Arizona Literature Mapper. Guided by cognitive fit and cognitive load theories, Chapter 3 develops a Web-based KMS, Nano Mapper, to support users' search and analysis of nanotechnology developments. The data sources include patent documents from world leading patent offices and NSF grand documents. Controlled experiments were conducted to evaluate the functions of Nano Mapper. The Web-based KMS developed in Chapters 4 and 5 is the Dark Web Forum Portal (DWFP) which aims to manage user-generated multilingual social media content. In Chapter 4, theories of the mental workload, task-technology fit, and unified theory of acceptance and use of technology (UTAUT) are used to guide the design and development of the system, and a new theoretical model is proposed to extend theory. Chapter 5 conducts two evaluation studies to comprehensively evaluate the performance of the system. The first study compared its performance with the benchmark system based on efficiency, effectiveness, and all the dimensions in the D&M IS Success Model. The second study further tested and validated the D&M IS Success Model in social media context.Overall, these research studies in my dissertation contribute to the literature on system development, knowledge mapping, technology acceptance and adoption, and human cognition and decision making.