The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
A intrusão é um dos principais problemas de segurança da Internet com o rápido crescimento de dispositivos inteligentes e de Internet das Coisas (IoT), e torna-se importante detectar ataques e acionar alarmes em sistemas IoT. Neste artigo, o método baseado em máquina de vetores de suporte (SVM) e análise de componentes principais (PCA) é usado para detectar ataques em sistemas IoT inteligentes. SVM com esquema não linear é usado para classificação de intrusão e PCA é adotado para seleção de recursos nos conjuntos de dados de treinamento e teste. Experimentos no conjunto de dados NSL-KDD mostram que a precisão do teste do método proposto pode chegar a 82.2% com 16 recursos selecionados do PCA para classificação binária, o que é quase igual ao resultado obtido com todos os 41 recursos; e a precisão do teste pode atingir 78.3% com 29 recursos selecionados do PCA para classificação múltipla, enquanto 79.6% sem seleção de recursos. A precisão da detecção de ataques de negação de serviço (DoS) do método proposto pode atingir uma melhoria de 8.8% em comparação com o método existente baseado em rede neural artificial.
Fei ZHANG
Northwestern Polytechnic University
Peining ZHEN
Shanghai Jiao Tong University
Dishan JING
Shanghai Jiao Tong University
Xiaotang TANG
Shanghai Jiao Tong University
Hai-Bao CHEN
Shanghai Jiao Tong University
Jie YAN
Northwestern Polytechnic University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Fei ZHANG, Peining ZHEN, Dishan JING, Xiaotang TANG, Hai-Bao CHEN, Jie YAN, "SVM Based Intrusion Detection Method with Nonlinear Scaling and Feature Selection" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 5, pp. 1024-1038, May 2022, doi: 10.1587/transinf.2021EDP7184.
Abstract: Intrusion is one of major security issues of internet with the rapid growth in smart and Internet of Thing (IoT) devices, and it becomes important to detect attacks and set out alarm in IoT systems. In this paper, the support vector machine (SVM) and principal component analysis (PCA) based method is used to detect attacks in smart IoT systems. SVM with nonlinear scheme is used for intrusion classification and PCA is adopted for feature selection on the training and testing datasets. Experiments on the NSL-KDD dataset show that the test accuracy of the proposed method can reach 82.2% with 16 features selected from PCA for binary-classification which is almost the same as the result obtained with all the 41 features; and the test accuracy can achieve 78.3% with 29 features selected from PCA for multi-classification while 79.6% without feature selection. The Denial of Service (DoS) attack detection accuracy of the proposed method can achieve 8.8% improvement compared with existing artificial neural network based method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2021EDP7184/_p
Copiar
@ARTICLE{e105-d_5_1024,
author={Fei ZHANG, Peining ZHEN, Dishan JING, Xiaotang TANG, Hai-Bao CHEN, Jie YAN, },
journal={IEICE TRANSACTIONS on Information},
title={SVM Based Intrusion Detection Method with Nonlinear Scaling and Feature Selection},
year={2022},
volume={E105-D},
number={5},
pages={1024-1038},
abstract={Intrusion is one of major security issues of internet with the rapid growth in smart and Internet of Thing (IoT) devices, and it becomes important to detect attacks and set out alarm in IoT systems. In this paper, the support vector machine (SVM) and principal component analysis (PCA) based method is used to detect attacks in smart IoT systems. SVM with nonlinear scheme is used for intrusion classification and PCA is adopted for feature selection on the training and testing datasets. Experiments on the NSL-KDD dataset show that the test accuracy of the proposed method can reach 82.2% with 16 features selected from PCA for binary-classification which is almost the same as the result obtained with all the 41 features; and the test accuracy can achieve 78.3% with 29 features selected from PCA for multi-classification while 79.6% without feature selection. The Denial of Service (DoS) attack detection accuracy of the proposed method can achieve 8.8% improvement compared with existing artificial neural network based method.},
keywords={},
doi={10.1587/transinf.2021EDP7184},
ISSN={1745-1361},
month={May},}
Copiar
TY - JOUR
TI - SVM Based Intrusion Detection Method with Nonlinear Scaling and Feature Selection
T2 - IEICE TRANSACTIONS on Information
SP - 1024
EP - 1038
AU - Fei ZHANG
AU - Peining ZHEN
AU - Dishan JING
AU - Xiaotang TANG
AU - Hai-Bao CHEN
AU - Jie YAN
PY - 2022
DO - 10.1587/transinf.2021EDP7184
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2022
AB - Intrusion is one of major security issues of internet with the rapid growth in smart and Internet of Thing (IoT) devices, and it becomes important to detect attacks and set out alarm in IoT systems. In this paper, the support vector machine (SVM) and principal component analysis (PCA) based method is used to detect attacks in smart IoT systems. SVM with nonlinear scheme is used for intrusion classification and PCA is adopted for feature selection on the training and testing datasets. Experiments on the NSL-KDD dataset show that the test accuracy of the proposed method can reach 82.2% with 16 features selected from PCA for binary-classification which is almost the same as the result obtained with all the 41 features; and the test accuracy can achieve 78.3% with 29 features selected from PCA for multi-classification while 79.6% without feature selection. The Denial of Service (DoS) attack detection accuracy of the proposed method can achieve 8.8% improvement compared with existing artificial neural network based method.
ER -