Efficient Harmful Email Identification Using NeuralNetwork

摘要

Phishing is a form of online fraud that aims to steal a user’s sensitive information such as online banking passwords or credit card numbers. In this paper, we present a technique to quickly detect suspicious email using Neural Network Pruning approach. The goal is to determine whether the email is suspicious or legitimate. A Multilayer feedforward neural network with Pruning Strategy is used for Feature Extraction and extracted features are used for identifying email as phishing email. Pruning Strategy extracts important features which are playing a key role in identifying phishing mail which looks similar to a legitimate one. To verify the feasibility of the proposed approach experimental evaluation has been performed using a dataset composed of phishing emails along with legitimate emails. The experimental results are satisfactory in terms of false positives and false negatives. The results of conducted test indicated good identification rate with very short processing time.