A Personalized Whitelist Approach for Phishing Webpage Detection

The number of phishing attacks against web services has seen a steady increase causing, for example, a negative effect on the ability of banking and financial institutions to deliver reliable services on the Internet. This paper presents an automatic approach detecting phishing attacks. Our approach combines a personalized whitelisting approach with machine learning techniques. The whitelist is used as filter that blocks phish web pages used to imitate innocuous user behavior. The phishing pages that are not blocked by the whitelist pass are further filtered using a Support Vector Machine classifier designed and optimized to classify these threats. Our experimental results show that the proposed approach improves over the current state-of-the-art methods.