A MACHINE LEARNING -BASED PHISHING ATTACK-RESISTANT MULTI-FACTOR DETECTION MODEL

Phishing remains one of the most pervasive and damaging threats in today’s digital landscape, often bypassing traditional detection systems through social engineering and technical obfuscation. This study proposes a robust and adaptive RMFDM to mitigate phishing attacks using machine learning techniques. To improve accuracy and resilience, the model integrates multiple detection layers, including URL analysis, domain reputation, content inspection, and behavioral features. Publicly available datasets were used to train and test machine learning classifiers such as SVM, decision tree, and KNN, with SVM yielding the highest performance. In addition to URL-based detection, the model incorporates biometric (face recognition) and behavioral (more code-based input) authentication mechanisms to further reinforce access control. The experimental results show that the multi-factor model achieves 98% accuracy, 96% precision, and 99% recall, significantly outperforming the traditional heuristic and single-layer detection approaches. The layered architecture of RMFDM ensures that even if one component is compromised, the other components maintain their integrity, making it resilient against zero-day and evolving phishing attacks. This study contributes a scalable, intelligent, and user-focused solution to phishing detection, with potential applications in secure authentication systems across web platforms, enterprises, and critical infrastructures