AI-Driven Fake Account Detection

This paper presents an AI-Driven Fake Account Detection System designed to identify fraudulent and suspicious accounts on social media platforms. The system combines behavioral, content, network, and metadata analysis to detect anomalies in user profiles. Machine learning techniques are used to evaluate account features such as activity patterns, engagement levels, follower-following relationships, and profile attributes. A risk scoring mechanism classifies accounts as genuine or fake with an associated confidence level. The system is developed using Python, Flask, Scikit-learn, and SQLite, providing a web-based interface for real-time analysis. Additional features including batch analysis, analytics dashboards, history tracking, and complaint reporting improve usability and effectiveness. Experimental results demonstrate reliable detection performance, making the system a practical solution for enhancing social media security and reducing fraudulent activities.