Phishing Detection Using Deep Learning and Machine Learning Algorithms: Comparative Analysis

Phishing attacks continue to pose a significant threat to online security, with attackers using increasingly sophisticated techniques to trick users into divulging sensitive information. In this paper, we compare the performance of two different Deep Learning (DL) models with three Machine Learning (ML) algorithms in detecting phishing attacks. The DL models include a combined model of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM), as well as a Multilayer Perceptron (MLP) model. Furthermore, the ML algorithms consist of Gradient Boosting Classifier (GBC), Logistic Regression (LR), and Naive Bayes (NB). By using a public dataset of more than 10,000 websites, our performance evaluation demonstrated that the combined DL model of CNN and LSTM outperformed all of the other models and algorithms used in this study, with an accuracy of 93.1%. On the other hand, the least-performing algorithm was NB, attaining a low accuracy of 66.0%.