Analysis on Detection of Spammers and Fake Users on Social Networks

Abstract

The widespread use of social networking platforms has significantly increased the risk of malicious activities, including the spread of spam, misinformation, and the creation of fake user profiles. This paper presents an analytical study on the detection of spammers and fake users on social networks using data mining and machine learning techniques. The proposed system analyzes user behavior patterns, content characteristics, and network interactions to distinguish between genuine and suspicious accounts. Key features such as posting frequency, friend-to-follower ratio, message similarity, and account activity are extracted and evaluated. Machine learning algorithms, including Decision Tree, Random Forest, and Support Vector Machine (SVM), are applied to classify users based on these behavioral attributes. The study highlights the importance of feature selection and model optimization to enhance detection accuracy. Experimental results using real-world social network datasets show that the proposed approach effectively identifies fake and spam accounts, improving platform security and user trust.