PROACTIVE BIMODAL SPAM COMMUNITY DETECTION IN ONLINE SOCIAL NETWORKS

In recent times, Online Social Networks like Twitter is facing more challenges in the direction of Spam accounts and spammer community. The popularity and open structure of twitter have attracted a large number of automated programs known as bots, which appear to be a double-edged sword to Twitter. Legitimate twitter users complain that their accounts are mistakenly suspended by twitter’s anti-spam action. Spammers are only identified after many benign users are affected. A comprehensive and empirical evaluation is performed in Twitter spam show that more and more spammers are evolving. The intention of the existing system is only to identify individual spam accounts. The spam accounts are classified using multi-class SVM classifier as namely benign, verified, spam, hacked accounts. In this paper, the suspicious spam community is without affecting the legitimate benign accounts using bimodal spam community detection algorithm whichinvolves graph network and content based features shows the effectiveness of the model