Software Engineering Institute

Cybersecurity Data Science (CSDS) encompasses the rapidly growing practice of applying data science to prevent, detect, and remediate cybersecurity threats. CSDS methods emerge from the application of data analytics and machine learning to challenges associated with security assurance. Starting with a brief introduction to data science, this presentation will introduce key CSDS use cases and challenges. Based on industry research and interviews with practitioners, an overview of key trends in the domain will be framed and discussed:

• Cloud-based cybersecurity analytics

• Real-time IoT/endpoint-based detection

• Deep learning and reinforcement learning

• Human-in-the-loop cyclical machine learning

• Machine learning facilitated adversarial mechanisms

• Adversarial attacks on machine learning systems

• AI-driven fake news and disinformation campaigns

• Possibilities and dynamics of cyber-AI cold war

Concluding guidance will be provided for defenders on building CSDS capabilities, both individually and within organizations. The presentation summarizes learnings gained by the author while writing the new book ‘Cybersecurity Data Science: Best Practices in an Emerging Profession’ (Mongeau, 2021).