HYBRID DETECTION OF FAKE & CLONE TWITTER(X) PROFILES

Abstract

The rapid growth of social media platforms such as Twitter (now known as X) has led to an increase in fake and clone profiles, which are often used for spreading misinformation, scams, and malicious activities. Detecting such profiles is a challenging task due to their ability to mimic legitimate users by replicating profile information, images, and behavioral patterns. This project proposes a hybrid detection system that combines machine learning techniques with rule-based analysis to identify fake and clone Twitter (X) profiles effectively. The system analyzes multiple features including user metadata (followers, following ratio, account age), profile attributes (username similarity, profile picture duplication), and behavioral patterns (posting frequency, content similarity). A combination of supervised learning algorithms such as Random Forest and Support Vector Machines is used alongside similarity detection techniques to improve accuracy. The hybrid approach enhances detection performance by leveraging both statistical patterns and heuristic rules.