Enriching Trajectories with Semantic Data for a Deeper Analysis of Patterns Extracted

Geographical Information System (GIS) stores several types of data collected from several sources in varied format. Thus geo-databases generate day by day a huge volume of data from satellite images and mobile sensors like GPS, among these data we find in one hand spatial features and geographical data, and in other hand trajectories browsed by several moving objects in some period of time. Merging these types of data leads to produce semantic trajectory data. Enriching trajectories with semantic geographical information lead to facilitate queries, analysis, and mining of moving object data. Therefore applying mining techniques on semantic trajectories continue to proof a success stories in discovering useful and non-trivial behavioral patterns of moving objects. The objective of this paper is to envisage an overview of semantic trajectory knowledge discovery, and spatial data mining approaches for geographic information system. Based on analysis of various literatures, this paper proposes a concept of multi-layer system architecture for raw trajectory construction, trajectory enrichment, and semantic trajectory mining.