Computer Science majors Erin Taylor and Ian Frankenburg, along with Math major Eric Peshkin, spent ten weeks understanding how geometry and topology, in tandem with statistics and machine-learning, can aid in quantifying anomalous behavior in cyber-networks. The team was sponsored by Geometric Data Anaytics, Inc., and used real anonymized Netflow data provided by Duke’s Information Technology Security Office.
The team produced features measuring cyber-behavior at the node, aggregate node, edge, and subnetwork level. Using both Python and MATLAB, they constructed tools that enabled the fitting of probabilistic models to sets of these features, and built visualization devices for these models.
Download the Executive Summary (PDF)
Client
- John Harer, Geometric Data Analytics
Project Manager
- Joe Marion, Statistics
Participants
- Erin Taylor, Duke University Computer Science
- Eric Peshkin, Duke University Mathematics
- Ian Frankenburg, Duke University Computer Science
