Human Behavior Mining: A Framework for Theorizing About mHealth Behavior Using Digital Trace Data

Process mining is a set of analytical techniques aimed at gaining insights into business processes in organizations. Recently, information systems scholars have recognized its potential for analyzing human behavior through digital trace data. In this paper, we draw on the conceptual and technical analogies between business processes and human behavior to thoroughly investigate the application and transfer of process mining techniques to the analysis of human behavior. This analysis, called human behavior mining (HBM), is conceptualized as a four-part framework. To illustrate HBM’s research potential, we apply this framework in an mHealth scenario. We explore dynamic concepts proposed by social cognitive theory to analyze changes in physical activity behavior through digital trace data collected through a dedicated app. This exemplary application demonstrates that HBM can be used to empirically test previously unspecified and uncontested dynamic concepts in human behavior. It also highlights HBM’s suitability for health analytics, given the vast amount of health-related behavior data available through apps and wearables, and the direct connection between behavior and health-related outcomes. Our research provides a dynamic and temporal perspective on human behavior, showcasing the potential of HBM to enrich theoretical frameworks in IS research.