Overview
Explore a novel approach to cyber threat analysis in this 21-minute IEEE conference talk. Delve into the ShadeWatcher system, which utilizes recommendation-guided techniques and system audit records to enhance threat detection. Learn about knowledge graph building, connectivity modeling, and machine learning applications in cybersecurity. Examine the effectiveness and efficiency of this innovative method through comprehensive evaluations and case studies.
Syllabus
Intro
Cyber Threats Are Everywhere
Analyze Cyber Threat using System Auditing
Previous Approaches using Audit Records
Recommendation as a Similar Problem
Recommendation-guided Cyber Threat Analysis
Knowledge Graph Builder (cont.)
First-order Connectivity Modeling
Higher-order Connectivity Modeling
Learning to Cyber Threat Analysis
Evaluation
Effectiveness in Cyber Threat Detection
Study of Recommendation-guided Analysis
System Efficiency
Conclusion
Taught by
IEEE Symposium on Security and Privacy