基于无监督聚类算法的通信网络安全态势预测方法

A Method of Communication Network Security Situation Prediction Based on Unsupervised Clustering Algorithm

针对传统方法存在的网络安全预测精准度低的问题,提出一种基于无监督类算法的通信网络安全态势预测方法.首先阐述无监督聚类算法的原理,然后通过观察模型的外部特征确定设备是否正常,对主要数据进行离散型特征数值型的转化,并对转化后的数据进行归一化的处理,采用归一化的方法对两个极端值统一计算,从输入的网络流量中提取相关特征,实现通信网络安全态势的预测.实验结果表明,研究方法对网络安全态势的预测精准度更高,为通信网络的安全提供一种解决办法.

Aiming at the problem of low accuracy of network security prediction in traditional methods,a method of communication network security situation prediction based on unsupervised algorithm is proposed.Firstly,the principle of unsupervised clustering algorithm is described.Then,by observing the external characteristics of the model to determine whether the equipment is normal,the main data is transformed into discrete characteristic numerical type,and the transformed data is normalized.The unified calculation of the two extremes is carried out in the normalized way,and the relevant characteristics are extracted from the input network traffic to realize the communication network Prediction of network security situation.The experimental results show that the research method is more accurate in the prediction of network security situation,and provides a solution for the security of communication network.