Author : Mr.Pankaja Kumar Kandi

Phishing locales which hopes to take the exploited people private information by occupying them to surf a phony site page that looks like a true blue one is another kind of criminal acts through the web and its one of the particularly worries toward various territories including e-dealing with a record and retailing. Phishing site identification is genuinely a capricious and component issue including various parts and criteria that are not steady. We proposed an insightful model for distinguishing phishing website pages dependent on Extreme Learning Machine. Sorts of site pages are diverse as far as their highlights. Thus, we should utilize a particular website page highlights set to forestall phishing assaults. We proposed a model dependent on AI methods to distinguish phishing website pages. We have recommended some new guidelines to have effective element grouping. The model has 30 data sources and 1 yield. Right now, 10-overlap cross-approval test has been performed. The normal grouping precision estimated as 95.05%.

Keywords: Phishing, Extreme Learning Machine, Feature Classification.