Efficient Classifier for Detecting Spam in Social Networks through Sentiment Analysis

Reaching hundreds million of users, major social networks have become important target media for spammers. Social networks provide communication between people to share information through internet. The unbounded growth of content and users pushes the Internet technologies usage to certain limitations. To extract relevant content from the voluminous data in social network which is being a phenomenal task, because of its dynamic nature the participation is more complex than the static database. This work concentrates on using sentiment analysis and applying efficient classifier to detecting the spammer actions. The main objective of the proposed work is to analyze the sentiment differences between spammers and normal users to detect spam message from the unwanted sites. Further discuss the relationship between sentiment score and spam reviews establish an instant sequence combined with discriminative rules to detect spam store and spam review efficiently. The system applies efficient classification algorithms for detecting spam in better way after sentiment analysis. Experimental results on social media datasets show the advanced performance of the proposed framework by sentiment analysis of social spammer detection with cost effective manner.