基于支持向量机的通信网络异常流量数据挖掘方法  

A Data Mining Method for Abnormal Traffic in Communication Networks Based on Support Vector Machines

作　　者：劳雪松[1] LAO Xuesong(Anhui Police Officer Vocational College,hefei Anhui 230031,China)

机构地区：[1]安徽警官职业学院,安徽合肥230031

出　　处：《信息与电脑》2023年第12期197-200,共4页Information & Computer

摘　　要：传统方法对通信网络流量异常数据挖掘的精准度和效率较低,安全性不高。基于此,提出基于支持向量机通信网络异常流量数据挖掘方法并对该方法进行设计。首先,基于支持向量机对通信网络流量进行特征选择,利用支持向量机在通信网络流量异常挖掘中,选取一对一的构造方法进行类别分类。其次,通过统计频率法选择通信网络流量特征子集并列出大体流程图。再次,对通信网络流量异常特征聚类分析,先计算通信网络流量特征数据记录的距离,再建立通信网络流量特征聚类流程。最后,识别和挖掘通信网络流量异常数据,设计出通信网络流量异常判别模型,通过基于二分法的通信网络流量数据特征分析和基于支持向量机的判别后完成了通信网络流量异常的数据挖掘。将设计方法与传统方法和基于多尺度数据挖掘方法进行对比,得出该方法更具有优势。The traditional methods are less accurate and efficient for communication network traffic anomaly data mining and less secure.Based on this,support vector machine based communication network anomaly traffic data mining method is proposed and the method is designed.Firstly,the feature selection of communication network traffic based on support vector machine is carried out,and the one-to-one construction method is selected for category classification in communication network traffic anomaly mining using support vector machine.Secondly,a subset of communication network traffic features is selected and listed in the general flow chart by statistical frequency method.Again,the communication network traffic anomaly features are clustered and analyzed by calculating the distance of communication network traffic feature data records and then establishing the communication network traffic feature clustering process.Finally,identifying and mining communication network traffic anomaly data,the communication network traffic anomaly discrimination model is designed,and data mining of communication network traffic anomalies is completed by the communication network traffic data characterization based on dichotomy and after the discrimination based on support vector machine.The designed method is compared with the traditional method and the multi-scale data mining based method,and it is concluded that the method is more advantageous.

关 键 词：支持向量机 通信网络异常 数据挖掘 

分 类 号：TN915.08[电子电信—通信与信息系统]