Anomaly Detection Beyond the Research Setting: An exploration of the use of statistics and machine learning to detect cyber attacks

Anomaly Detection Beyond the Research Setting: An exploration of the use of statistics and machine learning to detect cyber attacks

Sæmundsson, G.D.

Hadziosmanovic, D. (mentor)
Asghari, H. (mentor)
Van Eeten, M.J.G. (mentor)

Technology, Policy and Management

Multi Actor Systems

SEPAM IA

2015-10-06

In this work we approach the problem of deploying anomaly detection techniques for detecting cyber attacks in an organisational environment. Anomaly detection has been an active research area for almost three decades with promising results. However, few such systems have been successfully im- plemented in an operational environment for improving cyber security. Researchers have attempted to identify the reasons for this gap between research and operational success, and provide guidelines on how to overcome it. In this work we use these guidelines to guide us in the exploration of how business organisations approach anomaly detection. We compare the insights from practice with theory in an effort to better understand the main discrepancies between the two settings.

anomaly detection
cyber security
intrusion detection
usability
organisational challenges

http://resolver.tudelft.nl/uuid:8b3cc7b9-e2c0-43c4-92cf-b8bde8cbbedf

Student theses

master thesis

(c) 2015 Sæmundsson, G.D.