基于机器学习的入侵检测方法研究综述

A Survey on Machine Learning Approaches for Developing Intrusion Detection System

入侵检测系统(Intrusion Detection System,IDS)主要用于检查主机或系统的活动,保护系统和数据框架免受恶意攻击.IDS能够跟踪网络中运行的硬件和软件的状态,以防恶意活动窃取数据.机器学习模型的应用可以使得IDS获得较低的误报率和较高的识别率.机器学习方法能够智能地识别正常和恶意流量,具有较高的识别精度.基于此,笔者重点介绍了多种基于机器学习的侵检测系统,通过对现有文献的广泛研究和调查,可以改进和创建高效入侵检测系统。

Intrusion detection system is mainly used to protect the system and data frame from malicious attacks.It checks the activity of the host or system.IDS tracks the status of hardware and software running in the network to prevent malicious activities from stealing data.The application of machine learning model can obtain lower false alarm rate and higher recognition rate.Machine learning method can intelligently identify normal and malicious traffic,and has high recognition accuracy.This paper focuses on the introduction of a variety of intrusion detection systems based on machine learning.Through extensive research and investigation of the existing literature,the gap between improving and creating efficient intrusion detection system can be determined.