Practical thinking on neural network phishing website detection research based on decision tree and optimal feature selection

Abstract

This article analyzes the detection methods of neural network phishing websites. The research content of this paper includes naive Bayes method, decision tree method, support vector machine method, neural network technology. The author combines the key points of phishing website detection based on decision tree and optimal feature selection to study such as URL feature and HTML feature analysis, website application feature analysis, K-Medoids cluster analysis, feature set screening. The author uses simulation experiments to complete the website performance check. The purpose of this article is to optimize the performance of phishing website detection and improve the security of the website's operating environment.