Information Security Network Intrusion Detection System Based on Machine Learning
Feng Guo, Hanlin Jiao, Xiong Zhang, Yuting Zhou, Hao Feng
Abstract
Nowadays, more and more people are becoming aware of the security issues of computers. Traditional IDS (Intrusion Detection System) has the drawbacks of poor real-time performance and low accuracy. In response to this issue, this article adopts a machine learning(ML) based network intrusion detection algorithm. The intrusion detection module can use SVM (Support Vector Machine), deep learning, etc., to construct models. The real-time detection and response module can use the established model to monitor and analyze network traffic in real-time, in order to achieve rapid detection and response to network attacks. In the real-time performance test results, the accuracy from 08:00 to 09:00 is 0.92, the false alarm rate is 0.05, and the average detection time is 50 milliseconds; The accuracy from 09:00 to 10:00 is 0.90, the false alarm rate is 0.06, and the average detection time is 55 milliseconds. This article studies a machine learning based method for detecting and identifying network attack behaviors, which is beneficial for improving the level of network security defense.