Social Media Spammer Detection by Unsupervised Techniques

Project Discription

Social media sites engage people by various means of social activities and public interest advisement. This gathering of people attract advertiser to increase their product reach with filtered audience. Social network is place to connect and share thoughts with each other. But most of people get attract from the social audience gathering for there personal or professional advantages. Social bots are social media accounts controlled completely or in part by computer algorithms. As social media want to maintain real users on platform so detection of spammers are proposed by this paper. This work has utilize soft computing technique, Intelligent water drop genetic algorithm for feature based user segregation. Users were segment into two classes first was spammer and other was real user. Proposed model do not need prior training, leaning, etc. as genetic algorithm IWD, dynamically segregate users. Experiment was done on twitter dataset and result shows that proposed has increase the precision value by 34.38% while accuracy of correct class identification was improved by 36.39%.

Project Sample Image