Mining Users’ Mobility Behaviors And Calculating User Similarity Based On Mobile Data

Mobile data contains a wealth of information. The information can not only reflect the user behavior. But also provides a lot of advanced information. Along with the increasing availability of location-acquisition and data mining technologies mining the potential information of mobile data has caused extensive concern of the researchers.Mobile data can reflect the users’ behavior. The behavior can be reflected by the different traffic modes. Mining users’ mobility behaviors can achieve through the identification of transportation modes. The traditional identification methods have the problem of low accuracy in a traffic jam or various transportation modes combined. In view of this situation, the combination of segmentation identification and supervised learning method was proposed. First, divide the original GPS trajectory into some sub-trajectories of single transportation mode by the data points, which speed is less than a threshold. Then the features of sub-trajectories are extracted. Use the supervised learning method to build an inference model to identify the transportation modes of different sub-trajectories. The experiment shows that the proposed algorithm can identify the different transportation modes effectively and achieve an ideal effect. At the same time this algorithm can also identify well in a traffic jam.Mobile data often exhibit certain temporal and spatial characteristics. Through the analysis of the spatial and temporal characteristics can be found with advanced information similar to the user. These data can also show the relationship between individuals, according to the relationship can provide some recommendation service. As a requirement for recommendation systems, the research of user similarity has also been widely concerned. To solve the user similarity between trajectories formed by mobility data, an algorithm based on Location Sequence Generalized Suffix Tree(LSGST) was proposed. First, the location sequence was extracted from mobility data. At the same time the location sequence was mapped to a string. The transformation from the processing of location sequence to the processing of string was completed. Then the location sequence generalized suffix tree between different users was constructed. The similarity was calculated in detail from the number of similar positions, longest common subsequence and the frequent common position sequence. The theoretical analysis and simulation results show that the proposed algorithm has ideal effect in terms of similarity measure. Besides, compared with the ordinary construct method, the algorithm has low time complexity. In the comparison with dynamic programming and naive string-matching, the algorithm has high efficiency in terms of similarity measurement and has low time complexity. Results of the simulation indicate that the LSGST can measure the similarity effectively, meanwhile reduces the trajectory data, and better performance in time complexity.