Contextual Security: A Critical Shift in Performing Threat Intelligence

Explore the critical shift towards contextual security in threat intelligence through this 23-minute conference talk from USENIX Enigma 2022. Delve into the challenges of automating trustworthy explanations of cyberattacks, examining the limitations of current machine learning-based security solutions and how contextual approaches can address them. Learn about the importance of deep system knowledge, real-time environmental data, and effective communication in cybersecurity. Discover the potential of knowledge graphs in threat intelligence and gain insights into ongoing research on explanation-based security. Understand key concepts such as continuous learning, provenance reasoning, and the generation of contextual knowledge graphs to enhance cybersecurity practices.

Introduction
What is Machine Learning
Limitations of Machine Learning
Continuous Learning
Knowledge Graphs
Generating a Knowledge Graph
Adding Context
Provenance Reasoning