Phishing Detection Using Data-Driven Intelligence

Since the early days of the Internet, phishing attacks have been a serious threat to the security of users and their personal information on the web. This problem raises the need to find a strict solution to minimize the cyber-attacks. This paper explores the application of machine learning models to detect phishing websites by analyzing multiple datasets containing phishing and legitimate website features. In order to evaluate machine learning models, in this study three publicly available phishing datasets were utilized. These datasets have different sizes and features. The performance of the model was evaluated on the basis of accuracy, precision, recall, and F1 score. The results showed that the Random Forest and Cubic SVM models achieved the highest accuracy for all datasets. Our findings highlight the importance of model selection and dataset characteristics in improving detection accuracy. Our future work will explore deep learning techniques to further improve phishing detection abilities.