Human navigation is ubiquitious: a soccer ball is passed between a sequence of players, when we navigate urban environments, we leave paths, and when we are searching in or interacting with information systems, our trails can be logged. Understanding the underlying generative processes of such sequences has implications, for example, for analyzing the strategy of a soccer team, optimizing urban traffic flows, or improving search experiences. In this talk, we introduce a versatile Bayesian framework for formulating and comparing hypotheses about the underlying generative processes of sequential processes and give examples of various application domains and the respective adaptations required to work with the corresponding data. Overall, the introduced method provides a comprehensive framework for analyzing and understanding sequential observations in a diverse field of different application domains.
