Universiti Teknologi Malaysia Institutional Repository

Using significant features for classification of attacks in intrusion detection

Zainal, Anazida and Maarof, Mohd. Aizaini and Shamsuddin, Siti Mariyam (2008) Using significant features for classification of attacks in intrusion detection. In: Advanced computer network & security. Penerbit UTM, Johor, pp. 127-146. ISBN 978-983-52-0613-9

[img] PDF (Abstract)
83Kb

Official URL: http://www.penerbit.utm.my/bookchapterdoc/FSKSM/bo...

Abstract

Finding good features to represent data is an important research area especially when dealing with problem domains that have many features such as bioinformatics, network intrusion detection (IDS) and many others. In IDS, accuracy and time are equally important. Unfortunately many reported works focused on getting high classification rate but fail to address the urgency of such detection. They use all the existing features in network traffic data to match against the known intrusive patterns. This has caused a lengthy detection process. Various techniques including machine learning and statistical approaches have been implemented and their detection accuracy is satisfactory. Among them are Artificial Neural Network [1-3], Support Vector Machine (SVM)[1][4-5], Bayesian Network and few others. Realizing the needs to uncover only the meaningful features from the abundant data, research in finding best feature subset has been intensified since early 2000. Both statistical and machine learning approaches were popularly used. [6] used Bayesian Network and Classification and Regression Tree, [7-8] used Flexible Neural Tree and few others have used other types of machine learning techniques.

Item Type:Book Section
Uncontrolled Keywords:network intrusion detection (IDS), intrusion detection
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:28138
Deposited By: Fazli Masari
Deposited On:07 Sep 2012 02:23
Last Modified:07 Sep 2012 02:23

Repository Staff Only: item control page