User Feature Recognition Based On Spatio-temporal Word Embedding Of Trajectory

The trajectory data generated from users' mobile access to base stations reflect their life styles and behavior patterns both in time and space.Based on the fact that temporal and spatial information are produced simultaneously,this paper proposes a TFT-IDFT method to extract semantic information from trajectories.First,a word embedding method named word2 vec is applied to build trajectory word vectors which include users' geometric and semantic information.Then,classification methods are used on these vectors to discriminate user age groups.The result shows that TFT-IDFT is more applicable than TF-IDF in the task of extracting semantic trajectories,and word vectors based on this method performs better in the age classification task.And K-Means method is used to do cluster analysis of 4G users and then analyze the users' characteristics of each category.The result shows that users can be divided into four groups by using the spatio-temporal data of mobile Internet,the features of these four groups are: crowds with high-repetition and narrow-scope of activities,crowds with high-repetition and wide-scope of activities,crowds with low-repetition and narrow-scope of activities,and crowds with low-repetition and wide-scope of activities.The difference of activities between workday and holiday is used to infer whether users from different groups doing that because of laziness or objective conditions.