Detection of Phishing Websites Using Machine Learning

Phishing is defined as imitating a creditable company's website aiming to take private information of a user. These phishing websites are to obtain confidential information such as usernames, passwords, banking credentials and some other personal information. Website phishing is the act of attracting unsuspecting online users into revealing private and confidential information which can be used by the phisher in fraud, blackmail or other ways to negatively affect the users involved. In this research, an approach had been proposed to detect phishing websites by applying a different kind of algorithms and filters to achieve a reliable and accurate result. The experiments were performed on four machine learning algorithms, e.g., SMO, logistic regression and Naive Bayes. Logistic regression classifiers were found to be the best classifier for the phishing website detection. In addition, the accuracy was enhanced when the filter had been applied to logistic regression algorithm.