DASHTrails: An Approach for Modeling and Analysis of Distribution-Adapted Sequential Hypotheses and Trails

The analysis of sequential trails and patterns is a prominent research topic. However, typically only explicitly observed trails are considered. In contrast, this paper proposes the DASHTrails approach that enables the modeling and analysis of distribution-adapted sequential trails and hypotheses. It presents a method for deriving transition matrices given a probability distribution over certain events. We demonstrate the applicability of the proposed approach using real-world data in the mobility domain, i.e., car trajectories and spatio-temporal distributions on car accidents.