Evolutionary learning program's behavior in neural networks for anomaly detection

Sang Jun Han, Kyung Joong Kim, Sung Bae Cho

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Learning program's behavior using machine learning techniques based on system call audit data is effective to detect intrusions. Among several machine learning techniques, the neural networks are known for its good performance in learning system call sequences. However, it suffers from very long training time because there are no formal solutions for determining the suitable structure of networks. In this paper, a novel intrusion detection technique based on evolutionary neural networks is proposed. Evolutionary neural networks have the advantage that it takes shorter time to obtain superior neural network than the conventional approaches because they learn the structure and weights of neural network simultaneously. Experimental results against 1999 DARPA IDEVAL data confirm that evolutionary neural networks are promising for intrusion detection.

Original languageEnglish
Pages (from-to)236-241
Number of pages6
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3316
Publication statusPublished - 2004 Dec 1

Fingerprint

Evolutionary Neural Networks
Evolutionary Learning
Anomaly Detection
Neural Networks
Intrusion Detection
Neural networks
Machine Learning
Learning systems
Formal Solutions
Audit
Intrusion detection
Learning Systems
Experimental Results

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{c6dcad06b8634bd682947b7663b9572d,
title = "Evolutionary learning program's behavior in neural networks for anomaly detection",
abstract = "Learning program's behavior using machine learning techniques based on system call audit data is effective to detect intrusions. Among several machine learning techniques, the neural networks are known for its good performance in learning system call sequences. However, it suffers from very long training time because there are no formal solutions for determining the suitable structure of networks. In this paper, a novel intrusion detection technique based on evolutionary neural networks is proposed. Evolutionary neural networks have the advantage that it takes shorter time to obtain superior neural network than the conventional approaches because they learn the structure and weights of neural network simultaneously. Experimental results against 1999 DARPA IDEVAL data confirm that evolutionary neural networks are promising for intrusion detection.",
author = "Han, {Sang Jun} and Kim, {Kyung Joong} and Cho, {Sung Bae}",
year = "2004",
month = "12",
day = "1",
language = "English",
volume = "3316",
pages = "236--241",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Evolutionary learning program's behavior in neural networks for anomaly detection

AU - Han, Sang Jun

AU - Kim, Kyung Joong

AU - Cho, Sung Bae

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Learning program's behavior using machine learning techniques based on system call audit data is effective to detect intrusions. Among several machine learning techniques, the neural networks are known for its good performance in learning system call sequences. However, it suffers from very long training time because there are no formal solutions for determining the suitable structure of networks. In this paper, a novel intrusion detection technique based on evolutionary neural networks is proposed. Evolutionary neural networks have the advantage that it takes shorter time to obtain superior neural network than the conventional approaches because they learn the structure and weights of neural network simultaneously. Experimental results against 1999 DARPA IDEVAL data confirm that evolutionary neural networks are promising for intrusion detection.

AB - Learning program's behavior using machine learning techniques based on system call audit data is effective to detect intrusions. Among several machine learning techniques, the neural networks are known for its good performance in learning system call sequences. However, it suffers from very long training time because there are no formal solutions for determining the suitable structure of networks. In this paper, a novel intrusion detection technique based on evolutionary neural networks is proposed. Evolutionary neural networks have the advantage that it takes shorter time to obtain superior neural network than the conventional approaches because they learn the structure and weights of neural network simultaneously. Experimental results against 1999 DARPA IDEVAL data confirm that evolutionary neural networks are promising for intrusion detection.

UR - http://www.scopus.com/inward/record.url?scp=35048873330&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048873330&partnerID=8YFLogxK

M3 - Article

VL - 3316

SP - 236

EP - 241

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -