摘要
随着入侵者的攻击手段日趋智能化、复杂化,传统的机器学习技术对异常攻击行为的检测有效性在下降。近年来,深度学习以其独特的学习机制,利用大数据和高算力达到学习的高准确率。通过广泛的文献调查,目前已经有很多基于深度学习设计的入侵检测系统。本综述在对传统机器学习技术和深度学习技术进行对比后,详述了基于深度学习和数据集的入侵检测系统。
As the attack methods of intruders become more intelligent and complicated,the effectiveness of traditional machine learning technology in detecting abnormal attack behaviors is declining.In recent years,deep learning uses big data and high computing power to achieve high learning accuracy with its distinctive learning mechanism.Through extensive literature surveys,many researchers have designed intrusion detection systems based on deep learning.This paper compared the traditional machine learning and deep learning,and detailed the intrusion detection systems based on deep learning and dataset.
作者
叶倩
谭天
孙艳杰
YE Qian;TAN Tian;SUN Yanjie(Hangzhou DPtech Information Technologies Co.,Ltd.,Hangzhou Zhejiang 310051,China)
出处
《信息安全与通信保密》
2021年第8期96-104,共9页
Information Security and Communications Privacy
关键词
入侵检测
机器学习
深度学习
数据集
intrusion detection
machine learning
deep learning
dataset