Research On Key Technologies Of Spatio Temporal Data Analysis In Mobile Social Networks

With the rapid development of wireless communication and mobile computing technology,the application of GPS and Beidou global positioning navigation system,convenient location acquisition methods has spawned a large number of moving objects(such as people,vehicles and animal)trajectory data.With the rapid growth of social network services market and fierce competition,current social networking services tend to move to form a mobile social network.In mobile social networks,people with common interests can communicate with each other using a mobile phone or tablet computer and other mobile devices.Users can track and share the location related information,massive spatial-temporal data in mobile social networks brings new opportunities for the study of mobile behavior in mobile social networks.In this paper,we propose a novel method for user mobility modeling and similarity measurement in mobile social networks,which aims at the requirement of massive spatial-temporal data analysis in mobile social networks.In addition,a new method based on Hausdorff distance is proposed to detect abnormal trajectories in the field of real-time security monitoring.In this paper,the real life dataset GeoLife,which is released by Microsoft Asia Research Institute,is used to evaluate the effectiveness and real-time performance of the proposed algorithm.The experimental results show that the proposed algorithm can effectively construct the mobile user's mobility model and the accuracy of similarity measurement is better than traditional methods.Moreover,trajectory anomaly detection algorithm proposed in this paper can effectively detect abnormal behaviors and reduce the time consumption of the traditional anomaly detection algorithm,which can make a certain contribution to the realization of online real-time anomaly detection.