基于小样本挖掘的网络安全漏洞识别研究  

Research on Network Security Vulnerability Identification Based on Small Sample Mining

作　　者：蔡秋富 刘海昇 田燕飞 CAI Qiufu;LIU Haisheng;TIAN Yanfei(Zhongkai College of Agricultural Engineering,Guangzhou 510000,China)

机构地区：[1]仲恺农业工程学院现代教育技术中心,广东广州510000

出　　处：《长江信息通信》2025年第1期156-158,共3页Changjiang Information & Communications

摘　　要：为提高网络用户信息与隐私信息的安全性,引进小样本挖掘技术,设计网络安全漏洞识别方法。运用支持向量机(Support Vector Machine,SVM)算法将样本划分间距较大、较小的样本,引入松弛因子,调整分类边界;引入主成分分析法,提取与聚类网络异常数据特征;构建网络安全态势评估模型,完成对网络安全态势计算与漏洞自动识别。实验结果表明:该方法可以实现对空间内正常数据与漏洞数据的准确区分。To improve the security of network user information and privacy information,small sample mining technology is introduced to design a network security vulnerability identification method.Using Support Vector Machine(SVM)algorithm to divide samples into larger and smaller intervals,introducing relaxation factor,and adjusting classification boundaries;Introduce principal component analysis to extract and cluster abnormal data features from the network;Build a network security situation assessment model to complete the calculation of network security situation and automatic identification of vulnerabilities.The experimental results show that this method can accurately distinguish between normal data and vulnerability data in space.

关 键 词：小样本挖掘 数据分类 漏洞识别 网络安全 

分 类 号：TN915[电子电信—通信与信息系统]