Spatio Temporal Data Patterns: Recent Results and Future Directions

by
Masud Hasan

In recent years, significant research has been done on analysis of
movement patterns in spatio-temporal data, where the primary objective
is to efficiently find and predict certain patterns among the large
amount of moving objects. Some well studied movement patterns are flock
(objects co-ordinately move close together), trend (anticipating future
movement of the objects), leadership (one or more objects spatially
leading a move of other objects), convergence (objects converge towards
a spot), and encounter (object meet at a spot). There exist many
practical applications of this research topic in diverse disciplines,
including GIS, database research, animal behavior research,
surveillance and security analysis, transportation analysis and market
research. In this talk, I shall survey the recent studies on this
research topic, with emphasis on designing efficient
algorithms and data structures for analyzing such patterns in
spatio-temporal data. I will also outline what can be done in future on
this topic.

Joint work with Alejandro Lopez-Ortiz (University of Waterloo)