An Ideal Approach for Detection and Prevention of Phishing Attacks

Abstract Phishing is a treacherous attempt to embezzle personal information such as bank account details, credit card information, social security number, employment details, and online shopping account passwords and so on from internet users. Phishing, or stealing of sensitive information on the web, has dealt a major blow to Internet security in recent times. These attacks use spurious emails or websites designed to fool users into divulging personal financial data by emulating the trusted brands of well-known banks, e-commerce and credit card companies. In this paper, we propose a phishing detection and prevention approach combining URL-based and Webpage similarity based detection. URL-based phishing detection involves extraction of actual URL (to which the website is actually directed) and the visual URL (which is visible to the user). LinkGuard Algorithm is used to analyze the two URLs and finally depending on the result produced by the algorithm the procedure proceeds to the next phase. If phishing is not detected or Phishing possibility is predicted in URL-based detection, the algorithm proceeds to the visual similarity based detection. A novel technique to visually compare a suspicious page with the legitimate one is presented.