Inferring social relationship in mobile social networks using tempo-spatial information

With the advancement of the Internet technology, network communication technology and mobile social applications, mobile social networks present the booming development momentum. Meanwhile, due to the popularity of mobile terminals such as smart phones and variety of PADs, it is feasible to obtain relatively accurate tempo-spatial information from mobile terminal holders. Therefore, inferring social relationships among those people on the basis of their historical tempo-spatial information becomes one of the hottest research fields. In this paper, we aim to infer social relationship among users according to their historical location information via the location-based service which is provided by many on-line social platforms. To achieve this goal, we firstly obtain historical location records and their corresponding timestamps which can be efficiently collected from selected on-line social platforms. Then, we infer the potential social relationship among users according to their historical tempo-spatial information by applying the social relationship inferring method proposed in this paper. Finally, by analyzing the experimental results, we make a set of helpful observations.