Semantically Enhanced Location-based Social Networks

Abstract

Trajectory data analysis has recently become an active research area. This is due to the large availability of mobile tracking sensors, such as GPS-enabled smart phones. However, those GPS trackers only provide raw trajectories (x, y, t), ignoring information about the geographical locations, transportation mode, etc. This information can contribute in producing significant knowledge about movements, which transforms raw trajectories into semantic trajectories. Therefore, research lately has focused on semantic trajectories; their representation, construction, and applications. Furthermore, advances in location acquisition and mobile technologies also led to the addition of the location dimension to Social Networks (SNs) and to the emergence of a newer class called Locationbased Social Networks (LBSNs). One of the key applications of semantic trajectories is location-based recommendation, which is a main function of LBSNs. This research investigates the current studies on semantic trajectories so far. We propose a new classification schema for the research efforts in semantic trajectory construction and applications. The proposed classification schema includes three main classes: semantic trajectory modeling, computation, and applications. Additionally we proposed a methodology to semantically enhance LBSNs through extracting SN Geo-tagged media annotations and using them as location semantics. This enabled