关键词: 知识图谱, 威胁情报, 受限玻尔兹曼机, 安全威胁感知, 威胁检测

Abstract: Aiming at the problems of complex sources, difficult to understand and share security threat intelligence, this paper realizes deep learning of threat intelligence features based on restricted Boltzmann machine, which maps the original threat intelligence features from high dimensional space to low dimensional space layer by layer, and constructs the cyberspace security threat knowledge map. By using the cyberspace security threat knowledge map, and combining with the current context, the path evolution and tracing of security threats are carried out through event flow processing to accurately perceive cyberspace security threats. The experiment verifies the feasibility of constructing the cyberspace security threat knowledge map, and verifies the security threat perception method based on the knowledge map is more suitable for the perception of highintensity security threats by comparing with traditional threat detection methods.

Key words: knowledge map, threat intelligence, restricted Boltzmann machine, security threat perception, threat detection