Performance Evaluation of Supervised Machine Learning Algorithms for Intrusion Detection

Manjula C. Belavagi, Balachandra Muniyal

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Intrusion detection system plays an important role in network security. Intrusion detection model is a predictive model used to predict the network data traffic as normal or intrusion. Machine Learning algorithms are used to build accurate models for clustering, classification and prediction. In this paper classification and predictive models for intrusion detection are built by using machine learning classification algorithms namely Logistic Regression, Gaussian Naive Bayes, Support Vector Machine and Random Forest. These algorithms are tested with NSL-KDD data set. Experimental results shows that Random Forest Classifier out performs the other methods in identifying whether the data traffic is normal or an attack.

Original languageEnglish
Pages (from-to)117-123
Number of pages7
JournalProcedia Computer Science
Volume89
DOIs
Publication statusPublished - 2016

Fingerprint

Intrusion detection
Learning algorithms
Learning systems
Network security
Support vector machines
Logistics
Classifiers

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

@article{d64a18e487dd4caa96af49cbfc7fbf3c,
title = "Performance Evaluation of Supervised Machine Learning Algorithms for Intrusion Detection",
abstract = "Intrusion detection system plays an important role in network security. Intrusion detection model is a predictive model used to predict the network data traffic as normal or intrusion. Machine Learning algorithms are used to build accurate models for clustering, classification and prediction. In this paper classification and predictive models for intrusion detection are built by using machine learning classification algorithms namely Logistic Regression, Gaussian Naive Bayes, Support Vector Machine and Random Forest. These algorithms are tested with NSL-KDD data set. Experimental results shows that Random Forest Classifier out performs the other methods in identifying whether the data traffic is normal or an attack.",
author = "Belavagi, {Manjula C.} and Balachandra Muniyal",
year = "2016",
doi = "10.1016/j.procs.2016.06.016",
language = "English",
volume = "89",
pages = "117--123",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

Performance Evaluation of Supervised Machine Learning Algorithms for Intrusion Detection. / Belavagi, Manjula C.; Muniyal, Balachandra.

In: Procedia Computer Science, Vol. 89, 2016, p. 117-123.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Performance Evaluation of Supervised Machine Learning Algorithms for Intrusion Detection

AU - Belavagi, Manjula C.

AU - Muniyal, Balachandra

PY - 2016

Y1 - 2016

N2 - Intrusion detection system plays an important role in network security. Intrusion detection model is a predictive model used to predict the network data traffic as normal or intrusion. Machine Learning algorithms are used to build accurate models for clustering, classification and prediction. In this paper classification and predictive models for intrusion detection are built by using machine learning classification algorithms namely Logistic Regression, Gaussian Naive Bayes, Support Vector Machine and Random Forest. These algorithms are tested with NSL-KDD data set. Experimental results shows that Random Forest Classifier out performs the other methods in identifying whether the data traffic is normal or an attack.

AB - Intrusion detection system plays an important role in network security. Intrusion detection model is a predictive model used to predict the network data traffic as normal or intrusion. Machine Learning algorithms are used to build accurate models for clustering, classification and prediction. In this paper classification and predictive models for intrusion detection are built by using machine learning classification algorithms namely Logistic Regression, Gaussian Naive Bayes, Support Vector Machine and Random Forest. These algorithms are tested with NSL-KDD data set. Experimental results shows that Random Forest Classifier out performs the other methods in identifying whether the data traffic is normal or an attack.

UR - http://www.scopus.com/inward/record.url?scp=84986536879&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986536879&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2016.06.016

DO - 10.1016/j.procs.2016.06.016

M3 - Article

VL - 89

SP - 117

EP - 123

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -