Phishing Detection in Websites Using Neural Networks and Firefly

Abstract

Nowadays phishing become popular in the internet. Phishing is a website forgery where the attackers steal sensitive information of users like username, password, bank details and security details without the knowledge of users. Phishers are the one to create website same as the trusted website with the same content and designs of the trusted website. Phishing can be done through email, websites and malicious software to get intellectual information, business secrets or military information etc. In order to prevent user from phishing websites PhishShield application is used. It detects phishing website with replacing content by images based on heuristic solutions. In this application an URL is given as input and it gives the status of URL whether it is legitimate or unknown or phishing websites. In this few features are used to detect phishing websites but in the proposed system we considered more features including Google PageRank, Google Position, Alexa rank and other URL based features and its accuracy and performance can be improved by using neural networks where optimum weight is calculated based on firefly algorithm. The experimental results are conducted to prove that the proposed technique works more effectively than the existing technique in terms of accuracy, true positive rate, true negative rate, false positive rate and false negative rate.