Comparison of Machine Learning Algorithms and Their Ensembles for Botnet Detection

Songhui Ryu, Purdue University

Abstract

A Botnet is a network of compromised devices controlled by a botmaster often for nefarious purposes. Analyzing network traffic to detect Botnet traffic has historically been an effective approach for systems monitoring for network intrusion. Although such system have been applying various machine learning techniques, little investigation into a comparison of machine algorithms and their ensembles has been undertaken. In this study, three popular classification machine learning algorithms—Naive Bayes, Decision tree, and Neural network—as well as the ensemble methods known to strengthen said classifiers are evaluated for enhanced results related to Botnet detection. This evaluation is conducted with the CTU-13 public dataset, measuring the training time and accuracy scores of each classifier.

Degree

M.S.

Advisors

Springer, Purdue University.

Subject Area

Information Technology

COinS