Fake Social Media Profile Detection and Reporting

Abstract: The rise of fake profiles on social media has triggered serious concerns around misinformation, scams, and digital harassment. Manual detection methods are inefficient at large scale, hence automated detection powered by machine learning is the need of the hour. This paper presents a real-time, machine learning- based detection system using the XGBoost classifier. The model analyzes behavioral and content-based features such as follower-following ratio, post frequency, profile bio characteristics, hashtag usage, and engagement patterns. The system is implemented through a user-friendly web interface using Streamlit, capable of flagging suspicious accounts with high accuracy. The solution aims to assist moderators and users in maintaining safe digital environments. Keywords: Fake Profiles, Machine Learning, social media, XGBoost, Streamlit, Behavioral Analysis, Detection System.