Identifying Phished Website Using Multilayer Perceptron

Abstract

Phishing is most popular in cybercrimes where a malicious individual or a group of individuals who scam users. The aim of identifying any phished website is to help the users/customers with more secure usage of online transactional websites. The research work focuses on the neural network concept which is implemented to identify phished websites. This concept is proved by multilayer perceptron (MLP)-based classification for 48 features. For result assessment, MLP is compared with other machine learning methods such as random forest, support vector machine (SVM), logistic regression and detected to have a higher accuracy of 96.80%.