基于逻辑回归模型的流量异常检测方法研究

发布时间：2018-06-02 14:02

本文选题：逻辑回归 + 机器学习　；参考：《工程数学学报》2017年05期

【摘要】：网络流量作为异常检测的基本数据源,其行为特征的准确描述,是网络异常行为实时检测的重要依据.本文针对流量异常检测问题,提出了一种基于逻辑回归模型的网络流量异常检测方法.通过分析源IP、目的 IP等多个网络流量基本特征,构造了网络异常行为和正常行为的训练机,并且在此基础上采用逻辑回归建立起网络异常流量挖掘模型.利用实验室所采集的真实网络流量对所构建的模型进行检测,以验证该模型的有效性.实验结果表明本文所建立的网络模型在异常流量挖掘方面准确度高、实时性好.
[Abstract]:As the basic data source of anomaly detection, the accurate description of the behavior characteristics of network traffic is an important basis for real-time detection of network abnormal behavior. In this paper, a network traffic anomaly detection method based on logical regression model is proposed. By analyzing the basic characteristics of network traffic, such as source IPand destination IP, a network abnormal behavior and normal behavior training machine is constructed, and a network abnormal traffic mining model is established by using logical regression. In order to verify the validity of the model, the real network traffic collected by the laboratory is used to detect the model. The experimental results show that the network model established in this paper has high accuracy and good real-time performance in abnormal traffic mining.
【作者单位】：西安理工大学高等技术学院;陕西省地方税务局;西北大学信息科学与技术学院;
【基金】：中国博士后科学基金(2014M560801) 陕西省自然科学基础研究计划项目(2014JQ8327)~~
【分类号】：TP393.06

【相似文献】