A Survey on the Detection of Phishing Websites

The number of network attacks is increasing at an unprecedented speed, and online phishing is the most common form of network attacks and the most close to our life. In order to enable users to better understand what kind of online phishing they have suffered, and enable scholars and researchers to more quickly customize detection for different types of online phishing detection. We need an in-depth summary of the existing types of online phishing and various detection methods. This paper classifies and summarizes all kinds of phishing, mainly introduces six kinds of phishing, including E-mail phishing and search engine phishing. For phishing detection methods, this paper also summarizes and classifies existing and widely used detection methods into seven categories such as list-based heuristic machine learning, and introduces them in detail. Then the paper analyzes the big data frontier technology and detection channel direction of assisted anomaly detection, and divides it into stream processing technology, Hadoop DPI parsing keyword recognition, web crawler recognition, etc. Finally the paper discusses and thinks deeply about the methods and techniques mentioned above as well as the current situation of phishing detection, and puts forward some forward-looking suggestions on the interpretation of data processing model selection and general detection architecture.