The Weibo Spammers' Identification and Detection based on Bayesian-algorithm
Yingying Huang, Mengyi Zhang, Yuqing Yang, Shijie Gan, Yanmei Zhang · 2016
The Weibo spammers are employed in latent network public relations or marketing companies, whose job is to achieve publicity or vilify the effect by means of organizing and planning the focus of speculation to a topic or person.In order to reduce the bad influence caused by them, this paper intends to establish a classifier based on the behavior characteristics.By analyzing the previous research, we set the ratio of followers, total number of blog posts, the number of friends, comprehensive quality evaluation and favorates according to latest data.Based on Bayesianalgorithm and Geneticalgorithm, we use R and Matlab to determine the optimal threshold matrix and conditional probability matrix of the changeable Weibo spammers.After testing, it has higher recognition accuracy.