摘要
对网络流量属性展开分析与选择是建立在流量特征评价标准模型的基础上,根据打包模型对网络流量进行改进。文章通过引入Relief算法,快速筛选具体权重,进而加快模型分类精准度与速度;通过K-Means聚类算法,文章建立针对网络流量监控识别的模型,并以此对相关网络平台数据包分类,根据已经具备的特征属性集对已知类型识别,在K-Means聚类算法应用下,查出未知类型。
The analysis and selection of network traffic attributes is based on the standard model of traffic characteristics evaluation,and the network traffic is improved according to the packaging model.By introducing Relief algorithm,this paper can quickly screen specific weights,and then accelerate the accuracy and speed of model classification.K-means clustering algorithm is used to establish a model for network traffic monitoring and identification,and then the data packets of the relevant network platforms are classified,and the known types are identified according to the existing feature attribute set.Under the application of the K-means clustering algorithm,the unknown types are found.
作者
易灿
彭婷
Yi Can;Peng Ting(Hunan Mass Media Vocational and Technical College,Changsha 410100,China)
出处
《无线互联科技》
2021年第17期108-109,共2页
Wireless Internet Technology
基金
湖南省教育厅科学研究一般项目,项目编号:20C0382。
关键词
网络流量特征
流量监控识别
网络流量分类
性能比较
network traffic characteristics
traffic monitoring and identification
network traffic classification
performance comparison