Comparative Analysis of Machine Learning and Deep Learning Techniques for Phishing Website Detection

Abstract

This study examines the performance of various machine learning and deep learning techniques for detecting phishing websites. Several classification models, including DenseNet, Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), XGBoost, and Multi-Layer Perceptron (MLP), are analyzed to determine their effectiveness in identifying fraudulent and potentially harmful websites. The experimental results indicate that XGBoost and MLP achieve the best classification performance, recording the highest accuracy among the evaluated models. DenseNet, a deep learning-based model, also provides promising results, with an accuracy of approximately 86%. Overall, the findings demonstrate that combining traditional machine learning algorithms with deep learning techniques can enhance the accuracy, reliability, and robustness of phishing website detection systems