Research And Implementation Of User Personality Analysis Based On Social Network

With the continuous development of the Internet,various kinds of Internet applications are becoming more and more popular.As an online content publishing and communication platform,social network has brought new changes to people's social life and communication mode.Because personality is highly correlated with behavior in cyberspace,the results of personality analysis can be applied to commodity recommendation system,personalized advertising and other scenarios,so acquiring personality of network users can further promote the development of related applications.Traditional personality measurement is mainly conducted by questionnaire,but this method is not suitable for large-scale users' personality measurement.Users will generate a lot of behavior data in the process of using social network platform.Through personality automatic prediction of social network users' behavior data,personality characteristics of social network users can be obtained more effectively.This thesis takes Facebook as the social network research platform,analyses the network behavior of Facebook users,and extracts the characteristics of user behavior,establishes the personality analysis model of social network users,and analyses the personality of Facebook users.The experiment verifies the feasibility of the model.The work done in this paper mainly includes the following points:(1)In the selection of user behavior characteristics,most previous studies only considered the statistical lexical features of user-published text,and ignored the intrinsic meaning of user-published text and the relationship between user psychological characteristics and personality.Relevant studies show that different personalities have certain differences in language habits,and the use of mental lexicon by different personalities is also different.Therefore,this topic carries out in-depth analysis of the attributes of users and the factors affecting users'behavior,distinguishes the differences between internal and external causes,puts forward text style features and psychological lexical features based on TF-IDF,and constructs a personality analysis model with multi-dimensional characteristics.(2)In the aspect of user behavior feature optimization,the binary particle swarm optimization(BPSO)algorithm is introduced,and adaptive mutation is carried out to solve the problem that BPSO is easy to fall into local optimum.The optimized BPSO is used for feature optimization,and the optimal feature combination is selected,which reduces the workload of feature extraction and improves the recognition efficiency and accuracy.(3)A user personality analysis system is designed an d implemented,which includes offline database of known tags,feature extraction module,personality analysis model training and testing module,user data collection and input module to be analyzed and user personality analysis module of social network.Experiments show that the systerm can effectively implement personality analysis of social network users.