摘要
传统恶意加密流量检测技术需要对其进行无差别解密,会增加隐私泄露的风险。而利用对流量特征进行的提取与整体分析思路构建的检测手段能够支持数据隐私保护功能。文章以此为研究对象,利用随机森林算法对流量特征进行识别,并在聚类分析的矩阵计算下提供对数据隐私保护的支持,可为此种流量检测确认方法的研究与应用奠定理论基础。
The traditional malicious encrypted traffic detection technology needs to decrypt it without discrimination,which will increase the risk of privacy disclosure.However,the detection method constructed by extracting traffic characteristics and overall analysis ideas can support data privacy protection functions.Therefore,this paper takes this as the research object,uses random forest algorithm to identify traffic characteristics,and provides support for data privacy protection under the matrix calculation of cluster analysis,hoping to lay a theoretical foundation for the research and application of this traffic detection and confirmation method.
作者
季奥颖
柳伟
丁页顶
鲍喜妮
高源
JI Aoying;LIU Wei;DING Yeding;BAO Xini;GAO Yuan(State Grid Zhejiang Electric Power Co.,Ltd.,Lishui Power Supply Company,Lishui 323000,China)
基金
2023年国网丽水供电公司群创储备项目“基于攻击链视角的全流量深度未知威胁检测研究与实现”(5211LS23000D)。
关键词
隐私保护
恶意加密
随机森林
检测
privacy protection
malicious encryption
random forest
detection
