PHISHING URL DETECTION: A REAL CASE SCENARIO THROUGH LOGIN URL’S

Abstract

Phishing is a common cybercrime in which attackers use fraudulent login URLs to steal usernames, passwords, banking credentials, and personal information. Increasingly sophisticated techniques such as URL masking, misleading domains, suspicious redirections, and evasion methods make traditional blacklist and signaturebased systems less effective against newly generated and zero-day phishing URLs. This project proposes a machine learning-based system for classifying real-world login URLs as legitimate or phishing. The framework collects genuine and phishing URLs and performs data preprocessing and feature extraction. Lexical and structural features such as URL length, domain characteristics, special characters, HTTPS usage, subdomains, and suspicious keywords are analyzed. Logistic Regression, Decision Tree, Random Forest, and Support Vector Machine (SVM) algorithms are trained and compared. Model performance is evaluated using Accuracy, Precision, Recall, F1- Score, ROC-AUC, and Confusion Matrix. The proposed system aims to provide accurate, efficient, and practical real-time phishing URL detection, reducing security risks during authentication and web browsing. Keywords: Phishing URL, Machine Learning, Logistic Regression, Decision Tree, Random Forest, SVM, Cybersecurity.