Detection of Trajectory Patterns and Visualization of Spatio-temporal Information Based on Data Stream Approaches

Abstract

With the rapid increase of the number of mobile GPS devices including smartphones, it is becoming more and more important to develop efficient and effective algorithms to analyze massive trajectory data streams generated through those devices. Although there are many algorithms that can find patterns from massive trajectory data stream by batch processes, what we need now is a new algorithm that can deal with massive data streams with limited resources by online processes. This study aims at developing such an algorithm and attempts to discover the places at which people often stop when they are walking or driving, or the places which are becoming crowded by analyzing massive trajectory data streams.