No abstract

No abstract

The numbers of online purchases and electronic banking transactions have increased substantially in the era of electronic business and mobile commerce. These online financial activities have attracted a special web threat called “phishing” that targets Internet users by seeking their credentials in order to access their financial information. Phishing involves impersonating a legitimate website by creating a visually similar fake website to deceive users. In the last decade different solutions to fight phishing that are primarily based on educating users, user’s experience, search methods, machine learning and features similarity have been developed. This paper combines computational intelligence along with user’s experience approaches to develop an anti-phishing visualisation method. Our method employs effective features chosen following thorough analysis on features scores generated by Correlation Feature Set and Information Gain processing techniques. We validate our anti-phishing features using classification systems produced by rule induction data mining approach. False positives, false negatives and phishing detection rate are the basis of evaluating the classification systems to measure our anti-phishing methods features’ integrity.