首页|基于信息挖掘的计算机通信网络异常流量节点检测算法

基于信息挖掘的计算机通信网络异常流量节点检测算法

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
为了实现计算机通信网络异常流量节点精准、有效检测，确保通信网络运行的安全性与稳定性，提出基于信息挖掘的计算机通信网络异常流量节点检测算法。通过分析计算机通信网络拓扑结构，判别通信网络实体的通信功能，辨识处理流量较大的关键网络节点;基于非相邻相减思路，对关键节点进行流量异常范围初步定位，输出异常节点判断阈值;利用K-means聚类算法对原始阈值进行更新，输出包含全部代表性特征的判断阈值，实现计算机通信网络异常流量节点检测。实验结果表明:利用设计方法所产生的节点检测准确率达到最高值为97%、拒检率达到最低值为6。8%，在计算数据量最大时的检测耗时为0。42 s，这说明设计方法的异常流量节点检测准确率较高、检测性能稳定、整体检测时间较短。
Information Mining Based Node Detection Algorithm for Anomalous Traffic Nodes in Computer Communication Networks
In order to realize the accurate and effective detection of abnormal traffic nodes in computer communication network,and ensure the safety and stability of communication network operation,the abnormal traffic node detection algorithm of computer communication network based on information mining is proposed.This paper analyzes the topology of computer communication network,discriminates the communication function of communication network entities,identifies the key network nodes with large traffic.This paper is based on the idea of non-adjacent phase reduction to preliminarily locate the abnormal traffic range of key notes and output the abnormal node judgment thresholds.This paper also uses the K-means clustering algorithm to update the original thresholds,and output the judgment thresholds that contain all representative features to achieve computer communication network abnormal traffic node detection.The experimental results show that the node detection accuracy rate generated by the design method reaches a maximum value of 97%,the rejection rate reaches a minimum value of 6.8%,and the detection time in the maximum amount of computational data is 0.42 s.This indicates that the design method has a higher accuracy rate of abnormal traffic node detection,stable detection performance,and a shorter overall detection time.

information mining techniquesK-means clustering algorithmnon-neighborhood subtractioncomputer communication networksanomaly traffic

关梅、冯宝珠、胡超

展开 >

蚌埠学院 数理学院,安徽 蚌埠 233030

安徽财经大学 工商学院,安徽 蚌埠 233030

安徽理工大学 空间信息与测绘工程学院,安徽 淮南 232001

信息挖掘技术 K-means聚类算法 非相邻相减 计算机通信网络 异常流量

国家自然科学基金资助项目安徽省高校自然科学研究重点项目蚌埠学院校级科学研究一般项目

11901510KJ2021A11312022ZR04

2024

新乡学院学报
新乡学院

新乡学院学报

影响因子：0.177
ISSN：2095-7726
年,卷(期)：2024.41(9)