Research And Implementation Of MITM Vulnerability Based On Machine Learning

With the rapid development of information technology,the Internet has become an indispensable part of people's life,but it brings high risk.Man-in-the-middle attack is a common means of hacking.It seriously affects user's information security.Therefore,the research of MITM is of great significance.Through analyzing the characteristics of MITM process,we build a system to detect MITM.According to the scene of MITM,This paper proposes a system based on network protocol and mode of exchanging information to get data.The system is mainly combined of active data module and passive data module.Active data module uses network tools such as OpenSSL,SSLScan and Namp to grasp the network configuration,such as communication protocol,certificate,communication secret key,session,cookie,etc.The passive data module mainly grabs the static web data,such as HTML,JS,forms,etc.This system provides data support for training detection model of MITM vulnerability.This paper presents a MITM vulnerability detection system based on machine learning.The system's training data come from MITM environmental analysis system.This paper presents a multi-source weighted model based on Bagging.In this paper,a large amount of data is used for testing to verifying the feasibility of model.Compared with the traditional model,this model has higher accuracy.