Heuristic based Approach for Phishing Site Detection Using URL Features

Abstract

Damage caused by phishing attacks that target personal user information is increasing. Phishing involves sending an email to a user or inducing a phishing page to steal a user’s personal information. This type of attack can be detected by blacklist-based detection techniques; however, these methods have some disadvantages and the numbers of victims have therefore continued to increase. In this paper, we propose a heuristic-based phishing detection technique that uses uniform resource locator (URL) features. We identified features that phishing site URLs contain. The proposed method employs those features for phishing detection. The technique was evaluated with a dataset of 3,000 phishing site URLs and 3,000 legitimate site URLs. The results demonstrate that the proposed technique can detect more than 98.23% of phishing sites. Keywords—phishing sites, URL-based features, heuristic, machine learning