BSides Brisbane - Beyond The Hype: Machine Learning for Security

•Anthony G. Tellez•2 min read
Machine LearningSecurityData ScienceRansomwareBotnetBSides

An overview of machine learning and AI concepts tailored for security analysts, cutting through the marketing hype to focus on practical applications and real-world use cases.

What You'll Learn

ML & AI Fundamentals

Understanding the core concepts:

  • What is data science?
  • The difference between ML and AI
  • The promise (and limitations) of AI for security analysts
  • Separating vendor marketing from practical reality

Practical Use Cases

Real-world applications with demonstrated results:

Ransomware Detection

Using machine learning to identify ransomware behavior patterns:

  • Behavioral analysis techniques
  • Feature engineering from endpoint data
  • Model training and validation
  • Operationalizing detection at scale

Botnet Detection

Applying ML to network data for botnet identification:

  • Network traffic analysis
  • Statistical anomaly detection
  • Command and control pattern recognition
  • Automated threat intelligence integration

Key Takeaways

Beyond the Marketing

Understanding what ML can and cannot do in security:

  • Realistic expectations vs. vendor claims
  • When to use ML vs. traditional approaches
  • Common pitfalls and how to avoid them

Operational Implementation

Making ML work in production environments:

  • Data requirements and quality
  • Model lifecycle management
  • Integration with existing security tools
  • Continuous improvement and feedback loops

The Security Data Scientist

What security practitioners need to know:

  • Essential ML concepts for security work
  • Tools and platforms available
  • Building vs. buying ML capabilities
  • Career paths in security data science

Presentation Materials

Download Slides

Presented at BSides Brisbane 2019

← Back to Blog