北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2020, Vol. 43 ›› Issue (2): 40-45.doi: 10.13190/j.jbupt.2019-094

• 论文 • 上一篇    下一篇

一种鲁棒网络流量分类及新类型发现算法

仇景明, 曲桦, 赵季红   

  1. 西安交通大学 信息与通信工程学院, 西安 710049
  • 收稿日期:2019-07-16 发布日期:2020-04-28
  • 通讯作者: 曲桦(1961-),男,教授,研究生导师,E-mail:qh@mail.xjtu.endu.cn. E-mail:qh@mail.xjtu.endu.cn
  • 作者简介:仇景明(1992-),男,硕士生.
  • 基金资助:
    国家自然科学基金项目(61531013)

A Robust Network Traffic Classification and New Type Discovery Algorithm

QIU Jing-ming, QU Hua, ZHAO Ji-hong   

  1. School of Information and Communication Engineering, Xi'an Jiaotong University, Xi'an 710049, China
  • Received:2019-07-16 Published:2020-04-28

摘要: 提出了一种鲁棒网络流量分类及新类型的发现算法.网络流量一般为高维数据,且在网络流量收集过程中易受到网络波动或网络攻击的影响,为此,在堆栈自编码器的基础上,基于互相关熵理论提出了一种新的网络模型进行数据的特征提取,通过基于阈值的主动学习分类算法进行分类,达到识别新应用类型的目的.对比实验结果表明,所提算法中分类算法的准确度可达到91.08%,对新应用类型的识别度可达到98.8%.

关键词: 稀疏自编码器, 特征提取, 主动学习, 鲁棒性, 新类型发现

Abstract: A robust network traffic classification and new type discovery algorithm is proposed by this paper, which is based on sparse autoencoder to extract feature features and classify based on threshold-based active learning classification algorithm. In addition, to achieve the purpose of identifying new application types, the excellent performance of the proposed algorithm through comparative experiments is verified. Among them, the accuracy of the classification algorithm can reach 91.08%; the recognition of new application types can reach 98.8%.

Key words: sparse self-encoder, feature extraction, active learning, robustness, new type discovery

中图分类号: