Online Program Home
  My Program

Abstract Details

Activity Number: 141 - Statistical Analysis of Cyber-Security Data
Type: Invited
Date/Time: Monday, July 31, 2017 : 10:30 AM to 12:20 PM
Sponsor: Royal Statistical Society
Abstract #322265 View Presentation
Title: Big Network Modeling and Anomaly Detection for Cyber-Security Applications
Author(s): Patrick Rubin-Delanchy*
Companies: University of Oxford
Keywords: cyber-security ; network ; anomaly detection ; central limit theorem ; hypothesis testing ; graph

Data arising in cyber-security applications often have a network, or `graph-like', structure, and accurate statistical modelling of connectivity behaviour has important implications, for instance, for network intrusion detection. We present a linear algebraic approach to network modelling, which is massively scalable and also very general. In this approach, nodes are embedded in a finite dimensional latent space, where common statistical, signal-processing and machine-learning methodologies are then available. A central limit theorem provides asymptotic guarantees on the statistical accuracy of the embedding. We explore an intriguing connection between `disassortivity', whereby nodes that are similar are relatively unlikely to connect, and space-time, as defined in special relativity. Mass testing for anomalous edges, correlations, and changepoints is then discussed. Results are illustrated on network flow data collected at Los Alamos National Laboratory. This is joint work with Nick Heard (Imperial College London).

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association