AI-DRIVEN SUPERVISED LEARNING MODEL FOR ADVANCED SPAM AND PHISHING DETECTION

Abstract

Email remains one of the most critical communication tools but also one of the most targeted by spam and phishing attacks that compromise privacy, data integrity, and organizational security. This paper presents an AI-driven supervised learning framework for detecting and classifying malicious email content. The model integrates advanced preprocessing methods, including tokenization, stop-word removal, and TFIDF feature extraction, with multiple supervised classifiers such as Support Vector Machine (SVM), Random Forest, and Logistic Regression. Experimental evaluations show that the hybrid model achieves superior precision, recall, and F1-scores compared to traditional filtering techniques. The inclusion of phishing detection alongside spam filtering broadens the system’s real-world applicability, enhancing digital safety and user trust in online communications.