APT attack path reconstruction method based on time sequence diagram comparison clustering and medium
By constructing a heterogeneous timing graph collection and combining threat intelligence knowledge graphs for graph clustering, the problems of insufficient event correlation and poor adaptability of complex attacks in the existing APT attack path reconstruction method are solved, and efficient and accurate APT attack path reconstruction is achieved.
Patent Information
- Application Number
- CN202510942976.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-09
- Publication Date
- 2025-08-12
- Estimated Expiration
- 2045-07-09
AI Technical Summary
The existing APT attack path reconstruction method is difficult to effectively express the correlation between events in large-scale, heterogeneous and fragmented security event streams, resulting in the multi-stage attack chain being segmented, lost or mis-stitched. The existing methods are poorly adaptable in complex concurrent attack activities, making it difficult to provide multi-dimensional visual replay methods, affecting response efficiency and accuracy.
By obtaining the standardized data set of security events, a heterogeneous timing graph collection is constructed and phase embedding annotation is performed, combining the multi-scale graph similarity algorithm of time alignment perception and threat intelligence knowledge graph, a semantic-guided weak supervision graph clustering framework is implemented, an attack path candidate graph is generated and integrity evaluation is performed, and logical conflicts are eliminated.
It improves the interpretability of APT traceability, reduces the subjective error of manual analysis, and improves the response efficiency and accuracy of attack path reconstruction.