Abstract

Cloned accounts, which mimic the identities of legitimate users, pose significant threats to privacy, brand reputation, and platform integrity. Detecting these impersonations is crucial for safeguarding users, maintaining trust within online communities, democracy, and election processes. This paper addresses the urgent need for effective detection mechanisms to combat the proliferation of these fraudulent profiles. By leveraging logistic regression, random forest, and XGBoost, we propose a robust framework for identifying cloned accounts on a collected dataset of 10,000 Twitter Accounts. Our approach demonstrates high accuracy rates, achieving 98.0%, 98.4% and 98.6% for Random Forest, XGBoost, and Logistic Regression, respectively, while also maintaining high F-measure scores. The study emphasizes the importance of analyzing user behavior, content patterns, and network features to enhance detection capabilities in real-time. The findings underscore the necessity for continuous innovation in detection systems to adapt to the evolving landscape of social media, where cloned accounts can engage in malicious activities such as misinformation dissemination and scams. Ultimately, this research contributes to the development of more reliable online identity verification systems, aiming to safeguard users and maintain the integrity of digital platforms.