Comparative analysis of features based machine learning approaches for phishing detection

Machine learning based anti-phishing techniques are based on various features extracted from different sources. These features differentiate a phishing website from a legitimate one. Features are taken from various sources like URL, page content, search engine, digital certificate, website traffic, etc, of a website to detect it as a phishing or non-phishing. The websites are declared as phishing sites if the heuristic design of the websites matches with the predefined rules. The accuracy of the anti-phishing solution depends on features set, training data and machine learning algorithm. This paper presents a comprehensive analysis of Phishing attacks, their exploitation, some of the recent machine learning based approaches for phishing detection and their comparative study. It provides a better understanding of the phishing problem, current solution space in machine learning domain, and scope of future research to deal with Phishing attacks efficiently using machine learning based approaches.