Network intrusion detection based on deep belief network and PSO-SVM

Bo Wang; Donggyun Kim; Kaoru Hirota; Yaping Dai

Network intrusion detection based on deep belief network and PSO-SVM

Bo Wang^*, Donggyun Kim, Kaoru Hirota, Yaping Dai

^*此作品的通讯作者

自动化学院

Beijing Institute of Technology

科研成果: 会议稿件 › 论文 › 同行评审

摘要

A hybrid algorithm for coupling PSO-SVM with Deep Belief Network (DBN) is applied to improve the recognition accuracy of intrusion detection. The hybrid algorithm has two parts, DBN is utilized to reduce the dimensionality of the feature sets. This is followed by a PSO-SVM to classify the intrusion into five outcome. The DBN reduces the dimension for the latency of data. And the PSO-SVM avoids a local optima. Simulation experiments using NSL-KDD dataset show that the proposed algorithm has higher detecting probability 93.83%, improve 1.66% compared with DBN-SVM.

源语言	英语
出版状态	已出版 - 2018
活动	8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Applications, ISCIIA and ITCA 2018 - Tengzhou, Shandong, 中国期限: 2 11月 2018 → 6 11月 2018

会议

会议	8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Applications, ISCIIA and ITCA 2018
国家/地区	中国
市	Tengzhou, Shandong
时期	2/11/18 → 6/11/18

其它文件与链接

链接到 Scopus 的出版物

引用此

@conference{047380c97d2345059b69a2cd2bc6063d,

title = "Network intrusion detection based on deep belief network and PSO-SVM",

abstract = "A hybrid algorithm for coupling PSO-SVM with Deep Belief Network (DBN) is applied to improve the recognition accuracy of intrusion detection. The hybrid algorithm has two parts, DBN is utilized to reduce the dimensionality of the feature sets. This is followed by a PSO-SVM to classify the intrusion into five outcome. The DBN reduces the dimension for the latency of data. And the PSO-SVM avoids a local optima. Simulation experiments using NSL-KDD dataset show that the proposed algorithm has higher detecting probability 93.83%, improve 1.66% compared with DBN-SVM.",

keywords = "Deep belief Network, Intrusion Detection, NSL-KDD, Particle Swarm Optimization with Support Vector Machine",

author = "Bo Wang and Donggyun Kim and Kaoru Hirota and Yaping Dai",

note = "Publisher Copyright: {\textcopyright} ISCIIA and ITCA 2018 - 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Application.; 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Applications, ISCIIA and ITCA 2018 ; Conference date: 02-11-2018 Through 06-11-2018",

year = "2018",

language = "English",

}

Wang, B, Kim, D, Hirota, K & Dai, Y 2018, 'Network intrusion detection based on deep belief network and PSO-SVM', 论文发表于 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Applications, ISCIIA and ITCA 2018, Tengzhou, Shandong, 中国, 2/11/18 - 6/11/18.

Network intrusion detection based on deep belief network and PSO-SVM. / Wang, Bo; Kim, Donggyun; Hirota, Kaoru 等.
2018. 论文发表于 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Applications, ISCIIA and ITCA 2018, Tengzhou, Shandong, 中国.

科研成果: 会议稿件 › 论文 › 同行评审

TY - CONF

T1 - Network intrusion detection based on deep belief network and PSO-SVM

AU - Wang, Bo

AU - Kim, Donggyun

AU - Hirota, Kaoru

AU - Dai, Yaping

N1 - Publisher Copyright: © ISCIIA and ITCA 2018 - 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Application.

PY - 2018

Y1 - 2018

N2 - A hybrid algorithm for coupling PSO-SVM with Deep Belief Network (DBN) is applied to improve the recognition accuracy of intrusion detection. The hybrid algorithm has two parts, DBN is utilized to reduce the dimensionality of the feature sets. This is followed by a PSO-SVM to classify the intrusion into five outcome. The DBN reduces the dimension for the latency of data. And the PSO-SVM avoids a local optima. Simulation experiments using NSL-KDD dataset show that the proposed algorithm has higher detecting probability 93.83%, improve 1.66% compared with DBN-SVM.

AB - A hybrid algorithm for coupling PSO-SVM with Deep Belief Network (DBN) is applied to improve the recognition accuracy of intrusion detection. The hybrid algorithm has two parts, DBN is utilized to reduce the dimensionality of the feature sets. This is followed by a PSO-SVM to classify the intrusion into five outcome. The DBN reduces the dimension for the latency of data. And the PSO-SVM avoids a local optima. Simulation experiments using NSL-KDD dataset show that the proposed algorithm has higher detecting probability 93.83%, improve 1.66% compared with DBN-SVM.

KW - Deep belief Network

KW - Intrusion Detection

KW - NSL-KDD

KW - Particle Swarm Optimization with Support Vector Machine

UR - http://www.scopus.com/inward/record.url?scp=85057969326&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85057969326

T2 - 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and Control Applications, ISCIIA and ITCA 2018

Y2 - 2 November 2018 through 6 November 2018

ER -

Network intrusion detection based on deep belief network and PSO-SVM

摘要

会议

其它文件与链接

指纹

引用此