摘要： 系统迁移上云、海量数据积累、智能技术应用等发展趋势给网络安全，特别是智能检测领域带来了诸多机遇和挑战。基于网络安全智能检测和隐蔽通信技术研究现状，从海量数据下的网络安全智能检测视角，提出了从数据采集到数据应用的五层框架和特征统计度量、分类模型算法等关键技术。在对网络隐蔽通信检测技术进行研究论述和讨论分析后，提出通过特征统计指标来评估数据流间的规律性、拟合度及相关性，设计了基于支持向量机（Support Vector Machine，SVM）的分类模型，并以分类特征向量为输入训练SVM分类器，实现对隐蔽通信的智能检测。

中图分类号：TP393.0；TP309文献标识码：ADOI:10.19358/j.issn.2097-1788.2025.04.005
引用格式：徐光亮，宿兴华，赵斯昂. 海量数据下的网络安全智能检测技术研究［J］.网络安全与数据治理，2025，44（4）：32-39.

Cyber security intelligent detection technology under massive data

Abstract： Development trends such as system migration to the cloud, massive data accumulation, and intelligent technology applications have brought many opportunities and challenges to cyber security, especially in the field of intelligent detection. Based on the research status of network security intelligent detection and covert communication technology, this paper proposes a 5-layer framework from data collection to data application, and key technologies such as feature statistical measurement and classification model algorithm from the perspective of network security intelligent detection under massive data. After researching and discussing the detection technology of network covert communication, this paper proposes to evaluate the regularity, fitting degree and correlation between data streams through characteristics statistics index, and designes a classification model based on support vector machine (SVM). The SVM classifier is trained with classification feature vector as input to realize the intelligent detection of covert communication.

Key words : cyber security; SVM; feature vectors; covert communication