Research On Social Network User Recognition Technology Based On Behavioral Characteristic Analysis

With the rapid development of network technology,social networks are playing an increasingly important role in people's daily lives.The wide availability and ease of use of social networks make it a platform for malicious users to spread spam.Although in some cases it is harmless for spammers to hide their true identities,in most cases it is for the purpose of deception.Real users of social platforms will receive spam messages from unknown sources,false advertising and promotion,which will eventually lead to phishing and fraud incidents and even trigger negative public opinion events on the Internet.Therefore,it is necessary to identify social network users to detect spam users.The research on social network user identification technology can effectively describe user characteristics and model social networks.This article mainly focuses on social network user identification technology based on behavior characteristics.Aiming at the current social network spam user identification method based on user behavior similarity,the main defect lies in the time performance problem when facing a large number of spam user groups.First,a candidate segmentation threshold algorithm is proposed to give the initial candidate segmentation threshold,linear detection is used to achieve candidate segmentation threshold approximation to the optimal segmentation threshold,so as to realize the social network spam user identification based on the optimal segmentation threshold.Relevant experimental results show that the method given in this paper has a great improvement in performance compared with existing methods when facing users with a large number of garbage groups.After analyzing the behavior sequence of social network users,it is found that complex user behaviors cause the coding sequence to be too long,resulting in large storage overhead.Based on this problem,by proposing the concept of compression threshold,a single continuous user behavior in the user behavior sequence encoding is compressed,so as to realize a social network user encoding technology based on an adaptive compression threshold.By compressing the user's behavior sequence,the storage overhead of coding-based user behavior characteristics is reduced.Relevant experimental results show that in the process of compressing some user behaviors with high similarity,this algorithm can effectively reduce the storage space required for user behavior sequences,and the effect of the spam group user detection method based on compression coding is also greatly improved.