CAJ | 학술논문

P2P流量逐渐成为互联网流量的重要组成部分,精确分类P2P流量对于有效管理网络和合理利用网络资源都具有重要意义.近年来,利用机器学习方法处理P2P流量分类问题已成为流量识别领域的一个新兴研究方向.利用决策树中的C4.5算法和P2P流量的特征属性来构建决策树模型,进而完成P2P流量分类问题.实验结果表明,基于决策树模型的方法能有效避免P2P网络流分布变化所带来的不稳定性;与SVM(support vector machine,支持向量机)、NBK(na(i)ve Bayes using kernel density estimation,改进的朴素贝叶斯)方法相比,其平均分类准确率能提高至少3.83个百分点.
P2P류량축점성위호련망류량적중요조성부분,정학분류P2P류량대우유효관리망락화합리이용망락자원도구유중요의의.근년래,이용궤기학습방법처리P2P류량분류문제이성위류량식별영역적일개신흥연구방향.이용결책수중적C4.5산법화P2P류량적특정속성래구건결책수모형,진이완성P2P류량분류문제.실험결과표명,기우결책수모형적방법능유효피면P2P망락류분포변화소대래적불은정성;여SVM(support vector machine,지지향량궤)、NBK(na(i)ve Bayes using kernel density estimation,개진적박소패협사)방법상비,기평균분류준학솔능제고지소3.83개백분점.
P2P traffic has become one of the most significant portions of the network traffic. Accurate identification of P2P traffic makes great sense for efficient network management and reasonable utility of network resources. In recent years, P2P traffic classification using machine learning has been a new direction in traffic identification. This paper proposed a new method based on decision-tree model, using C4.5 and P2P traffic characteristic. The experiments show this method can effectively avoid the instability of P2P traffic distribution change. Compared with SVM and NBK method, the average of classified precision can increase at least 3.83 percentage points.