Intelligent Phishing URL Detection with Hybrid ML Models

Abstract

Phishing attacks are one of the most prevalent and dangerous forms of cybercrime, exploiting email and websites to deceive individuals and steal sensitive information. With the increasing frequency of phishing incidents, robust defense mechanisms leveraging advanced machine learning techniques are essential. This study introduces a hybrid LSD (Logistic Regression, Support Vector Machine, and Decision Tree) model designed to enhance phishing detection accuracy and efficiency. The LSD model employs both soft and hard voting mechanisms to combine the strengths of the constituent algorithms. For feature selection, the canopy method is applied, followed by cross-validation and hyperparameter tuning using Grid Search to optimize performance. The effectiveness of the proposed system is evaluated using precision, accuracy, recall, F1- score, and specificity. Comparative analysis reveals that the hybrid LSD model significantly outperforms standalone classifiers, including Multinomial Naive Bayes, demonstrating its potential for proactive phishing defense.