使用机器学习算法分类P2P流量的方法研究  

Research on classification of P2P traffic using machine learning algorithm

作　　者：罗远军[1] LUO Yuanjun(School of Computer,Elctronics and Information,Guangxi University,Nanning 530004)

机构地区：[1]广西大学计算机与电子信息学院,南宁530004

出　　处：《计算机应用文摘》2022年第2期100-102,共3页Chinese Journal of Computer Application

摘　　要：随着时间的推移﹐P2P的应用数量、范围开始快速增长,这在促进社会经济发展和社会大众日常生活水平提升的同时,也带来了较多的问题,而传统的网络流量分类方式存在着较多的缺陷,难以满足时代的发展和社会大众的需求。文章主要提出基于ReliefF-CFS的方法,进而实现流量的特征子集的选择,依据具体的实验研究结果来看,文章所提出的方法取得了较高的分类准确率,比较适合在后续进行推广应用。因此,文章针对使用机器学习算法分类P2P流量的方法进行系统的研究和分析,其主要目的在于高效率处理协议加密的网络流量分类问题,这对于国内的计算机网络技术的发展也具有促进作用。With the passage of time,the number and scope of P2P applications began to grow rapidly,promoting social and economic development and improving the daily living standards of the general public,however,the traditional network traffic classification method has many defects,which can not meet the needs of the development of the times and the public.In this paper,a method based on relief-efs is proposed to select the characteristic subset of traffic.According to the experimental results,the method presented in this paper achieves a good classification accuracy,and it is more suitable for popularization and application in the follow-up.Therefore,in this paper,we will study and analyze the method of P2P traffic classification using machine learning algorithm,the main aim of which is to deal with the network traffic classification problem of protocol encryption efficiently.This also has the promoter action to the domestic computer network technology development.

关 键 词：机器学习算法 算法分类 P2P流量 智能算法 研究分析 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]