Outsmarting Phishers: A Comparative Analysis of Machine Learning Techniques

Phishing attacks threaten the security of the internet by stealing confidential data and money as well. As a way to prevent phishing, an extensive comparative study of the top most machine learning methods for phishing site detection was carried out. This research analyses the performance of ANN, RNN, XGBoost and Random Forest algorithms in the identification of phishing websites using the Kaggle dataset. These algorithms were selected due to their ability to uncover intricate associations and patterns from website information. The review examines the advantages and disadvantages each algorithm presents and compares them to each other based on accuracy efficient, precision, recall, F1 score, and computing efficiency. Through the comparison of these algorithms, the most effective algorithm for phishing detection is revealed, which can be useful to scholars and experts who focus on the improvement of on-line security. The research helps deposit the foundations for attacks prevention and facilitates the protection of online sensitive information. This study shows the effectiveness of using machine learning in the field of cybersecurity, especially with focus on the algorithms and how they can be optimized.