基于Random Forest与SVM算法的流量识别系统

Traffic Recognition System Based on Random Forest and SVM Algorithms

随着互联网的飞速发展,根据网络流量识别网络业务的类型,逐渐成为网络技术研究的重要课题。本文将SVM和Random Forest算法应用于流量识别系统的机器学习过程中,首先通过Random Forest算法对采集的数据特征信息进行分析选择,提取出在SVM算法中用来识别流量类型的8个主要特征,进而对数据进行预处理、训练学习,最终完成网络流量的分类识别。通过实验验证,该系统对流量识别准确率达96.7%,对当前的互联网应用的数据流量具有较高的识别准确率。

With the rapid development of the Internet,identifying the types of network services according to network traffic has gradually become an important topic of network technology research.In this paper,SVM and Random Forest algorithm are applied to the machine learning process of traffic identification system.Firstly,Random Forest algorithm is used to analyze and select the characteristic information of the collected data.Eight main features used to identify traffic types in SVM algorithm are extracted,and then the data are preprocessed,trained and learned.Finally,the classification and identification of network traffic is completed.The experimental results show that the accuracy of traffic identification reaches 96.7%,and the system has a high accuracy of data traffic identification for current Internet applications.