Using Ontology Based Semantic Association Rule Mining in Location Based Services

Abstract

Recently, GPS and mobile devices allowed collecting a huge amount of mobility data. Researchers from different communities have developed models and techniques for mobility analysis. But they mainly focused on the geometric properties of trajectories and do not consider the semantic facet of moving objects. The techniques are good at extracting patterns, but they are hard to interpret in a specific application domain. This paper proposes a methodology to understand mobility data and semantically interpret trajectory patterns. The process considers four different behavior types such as semantic, semantic and space, semantic and time, and semantic and space-time. Finally, a system prototype was developed to evaluate the behavior models in different aspects using one of the location based services. The results showed that applying the semantic association rules could significantly reduce the number of available services and customize the services based on the rules. KEYWORDS Data mining, ontology, semantic, location based services, association rule mining, spatiotemporal data