SuriCon 2017 - Hunting BotNets: Suricata Advanced Security Analytics

Anthony G. Tellez1 min read
Machine LearningSecuritySuricataBotnetConference

Splunk has enabled big data on the security practitioner's desktop, but the security knowledge worker is not a data scientist by training. SOC engineers need easy-to-implement machine learning tools.

This presentation explores existing machine learning toolkits available in the Splunk platform and how they can be applied to botnet hunting and advanced threat detection.

Key Topics

Botnet Detection

Using machine learning to identify botnet activity:

  • Command and control communication patterns
  • Periodic beaconing detection
  • Unusual traffic patterns
  • Behavioral analysis

Data Exfiltration

Detecting data theft through statistical analysis:

  • Volume anomalies
  • Destination analysis
  • Protocol anomalies
  • Time-based patterns

Port/Traffic Analysis

Network security analytics:

  • Port scanning detection
  • Traffic volume analysis
  • Protocol distribution
  • Connection pattern analysis

Advanced Threat Detection

Machine learning for sophisticated threats:

  • Zero-day detection
  • Polymorphic malware
  • APT activity identification
  • Behavioral baselines

Operationalizing ML

Making machine learning work in production SOC environments:

  • Model training workflows
  • False positive management
  • Integration with SIEM
  • Analyst collaboration

Presentation Materials

Download Slides


Presented at SuriCon 2017