Scam & Account Security

Abstract This project introduces an intelligent web-based system designed to detect fake and bot accounts on social media platforms. The system features a frontend built with TailwindCSS to deliver an interactive dashboard, and a backend powered by FastAPI for efficient data processing and analysis. Its primary purpose is to assist marketers and analysts in verifying account credibility and reducing metric inflation caused by non-authentic followers. The methodology transforms raw user data into behavioral indicators—such as engagement rate and posting-time variability—followed by the application of statistical techniques including Z-Score analysis and the Interquartile Range (IQR) method to identify abnormal patterns. The project addresses the growing challenge of detecting fraudulent accounts, offering an essential solution for enhancing data reliability in digital marketing. Results demonstrated that the system can effectively distinguish human behavior from bot-like activity by identifying statistical anomalies and unnatural temporal patterns, confirming its accuracy and reliability.