Visual Interactive Exploration of Spatio-Temporal Patterns

In many applications it is of high interest to analyze spatio-temporal (ST) datasets for recurring patterns. In our approach we examine spatio-temporal datasets for movement patterns using efficient query techniques. In an iterative process combining search and interaction an expert can review search results and either draw direct conclusions and annotate the pattern or refine the search pattern. In this way a hierarchical visualization / generalization and interactive analysis of large ST datasets becomes viable. The key feature of the approach is that patterns are not specified in advance (as meaningful semantic patterns) but established from the data set.