ML - Enhanced Website Phishing Detector

Abstract

Phishing attacks remain a major cybersecurity concern, exploiting user trust through deceptive emails, websites, and links to steal sensitive data such as passwords and financial details. Traditional detection systems often fail to identify new and sophisticated phishing patterns due to their dependency on predefined rules. To address this limitation, PhishGuard introduces a proactive detection framework that integrates a lightweight browser extension with a backend machine learning model. The system captures URLs and webpage features, performs intelligent feature extraction, and classifies potential threats using trained algorithms. When suspicious activity is detected, real-time alerts are generated to prevent user compromise. Experimental analysis shows that PhishGuard achieves high detection accuracy with minimal false positives, ensuring strong protection without affecting browser performance. This research contributes to the field of automated phishing detection by combining real-time monitoring, adaptive learning, and user transparency to enhance online security effectively.