Mobile Location Prediction in Spatio-Temporal Context

The increasing use of mobile devices and popular mobile services has led to massive availability of mobile data. Location prediction is a specific topic in mobile data mining, with its potential application in traffic planning, location-base advertisement, and user oriented coupon dispersion. Traditional location prediction methods often separately consider spatial or temporal approach. Although there have been some attempts to integrate both spatial and temporal information for location prediction, most of them suffer from the overfitting problem due to the large number of spatio-temporal trajectory patterns. Therefore, smoothing techniques are indispensable for proper training of spatio-temporal models. In this paper, we propose a novel location prediction model to capture the spatio-temporal context of user visits. It considers not only the spatial historical trajectories, but also the temporal periodic patterns. By applying smoothing techniques on both patterns, our model obtains significant improvement compared to the state-of-the-art approaches.