Integration of the rough set theory and support vector machines for the network intrusion detection

碩士 === 東吳大學 === 資訊管理學系 === 99 === Network attacks recently receive much attention because of the quickly development of businesses in the internet. Therefore, network intrusion detection has become an important issue in the area of network security. The goal of network security is to correctly ident...

Full description

Bibliographic Details
Main Authors: Wei-yen Shen, 沈威諺
Other Authors: Jih-Jeng Huang
Format: Others
Language:zh-TW
Published: 2011
Online Access:http://ndltd.ncl.edu.tw/handle/14179183008562149931
id ndltd-TW-099SCU05396024
record_format oai_dc
spelling ndltd-TW-099SCU053960242016-04-11T04:22:42Z http://ndltd.ncl.edu.tw/handle/14179183008562149931 Integration of the rough set theory and support vector machines for the network intrusion detection 整合約略集合理論與支援向量機於網路入侵偵測之應用 Wei-yen Shen 沈威諺 碩士 東吳大學 資訊管理學系 99 Network attacks recently receive much attention because of the quickly development of businesses in the internet. Therefore, network intrusion detection has become an important issue in the area of network security. The goal of network security is to correctly identify the attack or normal traffics. In this paper, the one-class support vector machine (one-class SVM) is first used to divide data flows into the non-outlier and outlier data. Then, the support vector machine (SVM) and rough sets theory (RST) are integrated to detect network intrusion. The support vector machine is used to classify the non-outlier data into the normal and attack data. On the other hand, the rough sets theory is used to divide the outlier data into the normal and attack data. If the outlier data could not be classified by the rough sets theory because they do not satisfy any rule, then we can assign the data to the attack data. In addition, we employed the KDD’99 dataset to justify the proposed method and to compare it with the conventional methods, such as support vector machine, logistic regression, discriminate analysis, artificial neural network (ANN) and classification and regression trees (CART). The experimental results shown that the accuracy of the proposed method is better than that of others. Jih-Jeng Huang 黃日鉦 2011 學位論文 ; thesis 59 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 東吳大學 === 資訊管理學系 === 99 === Network attacks recently receive much attention because of the quickly development of businesses in the internet. Therefore, network intrusion detection has become an important issue in the area of network security. The goal of network security is to correctly identify the attack or normal traffics. In this paper, the one-class support vector machine (one-class SVM) is first used to divide data flows into the non-outlier and outlier data. Then, the support vector machine (SVM) and rough sets theory (RST) are integrated to detect network intrusion. The support vector machine is used to classify the non-outlier data into the normal and attack data. On the other hand, the rough sets theory is used to divide the outlier data into the normal and attack data. If the outlier data could not be classified by the rough sets theory because they do not satisfy any rule, then we can assign the data to the attack data. In addition, we employed the KDD’99 dataset to justify the proposed method and to compare it with the conventional methods, such as support vector machine, logistic regression, discriminate analysis, artificial neural network (ANN) and classification and regression trees (CART). The experimental results shown that the accuracy of the proposed method is better than that of others.
author2 Jih-Jeng Huang
author_facet Jih-Jeng Huang
Wei-yen Shen
沈威諺
author Wei-yen Shen
沈威諺
spellingShingle Wei-yen Shen
沈威諺
Integration of the rough set theory and support vector machines for the network intrusion detection
author_sort Wei-yen Shen
title Integration of the rough set theory and support vector machines for the network intrusion detection
title_short Integration of the rough set theory and support vector machines for the network intrusion detection
title_full Integration of the rough set theory and support vector machines for the network intrusion detection
title_fullStr Integration of the rough set theory and support vector machines for the network intrusion detection
title_full_unstemmed Integration of the rough set theory and support vector machines for the network intrusion detection
title_sort integration of the rough set theory and support vector machines for the network intrusion detection
publishDate 2011
url http://ndltd.ncl.edu.tw/handle/14179183008562149931
work_keys_str_mv AT weiyenshen integrationoftheroughsettheoryandsupportvectormachinesforthenetworkintrusiondetection
AT chénwēiyàn integrationoftheroughsettheoryandsupportvectormachinesforthenetworkintrusiondetection
AT weiyenshen zhěnghéyuēlüèjíhélǐlùnyǔzhīyuánxiàngliàngjīyúwǎnglùrùqīnzhēncèzhīyīngyòng
AT chénwēiyàn zhěnghéyuēlüèjíhélǐlùnyǔzhīyuánxiàngliàngjīyúwǎnglùrùqīnzhēncèzhīyīngyòng
_version_ 1718221248336494592