The rapid growth of social media platforms has led to a significant increase in fake or malicious profiles. These fake accounts are often used for spreading misinformation, performing scams, spamming, and manipulating public opinion. Detecting such fake profiles has become an important challenge in social network security. This research proposes a machine learning-based approach for identifying fake profiles using user activity features such as number of friends, posts, likes, and comments. A synthetic dataset was generated using the Faker library to simulate real-world social media data. Data preprocessing techniques such as label encoding and feature scaling were applied before training the machine learning model. A Random Forest classifier was used to predict whether a profile is fake or genuine. Experimental results demonstrate that the proposed model achieves satisfactory accuracy in detecting fake accounts. The system also includes data visualization techniques for better analysis of user behavior. This research highlights the effectiveness of machine learning techniques in identifying suspicious social media profiles and contributes to improving online security.
