论文部分内容阅读
随着信息技术的发展,对等网络P2P信息流量经常出现偏离正常范围的异常情况,以决策树算法为基础,对P2P流量检测和流量异常时的检测技术进行了研究。采用改进的C4.5决策树P2P流量检测模型,通过P2P流量异常检测模型对大量训练数据集的训练,实现了对对错误的逐步修正,通过实验室仿真试验,经过选择网络流量特征后,基于改进的C4.5决策树的P2P网络流量分类器能实现较好的分类效果,分类检测率在94.6%~96.7%,较高的检测率说明采用改进的C4.5决策树算法能有效对P2P流量进行检测,为今后研究P2P流量异常检测技术提供了参考。
With the development of information technology, Peer-to-Peer (P2P) P2P traffic often shows abnormal departures from the normal range. Based on the decision tree algorithm, the P2P traffic detection and traffic abnormality detection techniques are studied. Based on the improved C4.5 decision tree P2P traffic detection model and the training of a large number of training datasets by the P2P traffic anomaly detection model, a correct correction of the errors is implemented. After the network traffic characteristics are selected based on the laboratory simulation, The improved C4.5 decision tree based P2P network traffic classifier can achieve better classification results with the detection rates of 94.6% ~ 96.7%. The higher detection rate indicates that the improved C4.5 decision tree algorithm is effective for P2P Traffic detection, P2P traffic anomalies for the future detection technology provides a reference.