ABSTRACT

This project outlines a method to circumvent the challenge and complexity of detecting and identifying e-banking phishing websites. Blacklist, Heuristic, visual similarity, and machine learning are just a few of the many methods used to stop phishing attacks. The most popular techniques are blacklists and heuristic evaluation methods since they are simple to use, but they are unable to identify new phishing attacks. Machine learning is now the most effective method for detecting phishing attacks since it can identify all of the shortcomings of other methods.

An effective method of identifying these phishing websites is through machine learning (ML) based models. The models, which are built on regression and classification of Data Mining methods, it is intelligent, reliable, and efficient. These algorithms are utilized to characterize and identify all the elements and regulations necessary to categorize the phishing website and the connections between them.

Machine learning algorithms are applied on the data which contains thousands of different types of URLs (defacement, benign, malware, phishing) (UCI Machine Learning Repository) the outcome shows that it is possible to identify e-banking phishing websites using machine learning approaches.