Trajectory pattern analysis for urban traffic

The increasing pervasiveness of location-acquisition tech- nologies (GPS, GSM networks, etc.) is leading to the collection of large spatio-temporal datasets and to the opportunity of discovering usable knowledge about movement behaviour, which fosters novel applications and services. In this paper, we apply a trajectory pattern extraction framework, called T-Pattern, to a real-world dataset, describing mobility of citizens within an urban area. The mining tool adopted is able to provide useful insights both in terms of common movements followed in the city, and, as by-product of the mining engine, in terms of spatial distribution and tempo- ral evolution of the traffic density. Both kinds of results are provided in the paper in a visual form, aimed at helping the analyst to better interpret them and link them to his/her existing background knowledge of the domain.