网络流量分类算法比较研究OA
Comparison Research on the Algorithms of Network Traffic Classification
准确的网络流量分类既是众多网络研究工作的重要基础,也是网络测量领域的研究热点.基于流特征的六种分类算法进行比较分析,实验结果表明,使用特征选择方法,SVM算法具有较高的整体准确率和较好的计算性能,适合用于网络流量分类.
Accurate traffic classification is of fundamental importance to numerous network activities and it has been a hot topic in network measurement for a long time. A comparison of six algorithms of traffic classification based on flow features is conducted. Analysis and experiment show that using feature seletion method the support vector machine (SVM) method has high accuracy and better computational performance for network traffic classification.
彭勃
安徽医科大学公共基础学院计算机系 合肥230032
信息技术与安全科学
网络流量分类机器学习特征选择
network traffic classification machine learning feature selection
《计算机与数字工程》 2012 (5)
12-14,3
安徽医科大学科研基金项目(编号:2010xkj040)资助.
评论