Human Decisions in User Modeling: Motivation, Procedure and Example Application

The aim of this study was to investigate the use and potential of the psychological theory of human-behavior modeling, called the Theory of Planned Behavior (TPB), in a user-modeling domain. We performed a user experiment involving a well-studied problem of user modeling, i.e., a recommender system (RS) for movies. As a part of the TPB, a survey to estimate the behavioral, normative and control beliefs regarding movie selection was designed. Using the participants’ responses an Ajzen model for movie genres was built and evaluated. An existing public dataset for context-aware movie recommendation, CoMoDa, was used to evaluate the proposed method. The results showed that the TPB approach led to an interesting explanation of movie genre selection. Among others, the potential applications of the TPB in recommender systems and the architecture of such a RS were addressed. Questions about the potential applications of the TPB in the user-modeling domain and its limitations and drawbacks were discussed.