Hacker intrusion detection based on text classification with word2vec model and convolutional neural network

Tongbo Wang; Kaoru Hirota; Chaojun Wang; Dongyun Kim; Yaping Dai

Hacker intrusion detection based on text classification with word2vec model and convolutional neural network

Tongbo Wang, Kaoru Hirota^*, Chaojun Wang, Dongyun Kim, Yaping Dai

^*Corresponding author for this work

School of Automation

Beijing Institute of Technology

Research output: Contribution to conference › Paper › peer-review

Abstract

A hacker intrusion detection method is proposed to tighten up Windows system security with text classification, which the Word2vec model and Convolutional Neural Network (CNN) are applied to distinguish hostile intrusion logs from a benign. The proposal makes full use of users’ operation commands contained in Windows system logs to analyze possible intrusion information and it supplies an application of text classification with CNN and word2vec on network security. For experiment, the dataset from a company having thousands of employees is converted to word vectors by word2vec model. The results of CNN with word embedding and classifier Support Vector Machine (SVM) with term frequency-inverse document frequency (TF-IDF) show the intrusion detection rates 95.56% and 87.55%, respectively. The proposed method can be integrated into antivirus software for hacker intrusion detection.

Original language	English
Publication status	Published - 2018
Event	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, China Duration: 2 Nov 2018 → 6 Nov 2018

Conference

Conference	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
Country/Territory	China
City	Tengzhou, Shandong
Period	2/11/18 → 6/11/18

Keywords

Convolutional Neural Network
Hacker Intrusion Detection
Support Vector Machine
Text Classification
Word2vec

Cite this

Wang, T., Hirota, K., Wang, C., Kim, D., & Dai, Y. (2018). Hacker intrusion detection based on text classification with word2vec model and convolutional neural network. Paper presented at 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, China.

Wang, Tongbo ; Hirota, Kaoru ; Wang, Chaojun et al. / Hacker intrusion detection based on text classification with word2vec model and convolutional neural network. Paper presented at 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, China.

@conference{cc212f310e9c4c7a9534db24b5c326bb,

title = "Hacker intrusion detection based on text classification with word2vec model and convolutional neural network",

abstract = "A hacker intrusion detection method is proposed to tighten up Windows system security with text classification, which the Word2vec model and Convolutional Neural Network (CNN) are applied to distinguish hostile intrusion logs from a benign. The proposal makes full use of users{\textquoteright} operation commands contained in Windows system logs to analyze possible intrusion information and it supplies an application of text classification with CNN and word2vec on network security. For experiment, the dataset from a company having thousands of employees is converted to word vectors by word2vec model. The results of CNN with word embedding and classifier Support Vector Machine (SVM) with term frequency-inverse document frequency (TF-IDF) show the intrusion detection rates 95.56\% and 87.55\%, respectively. The proposed method can be integrated into antivirus software for hacker intrusion detection.",

keywords = "Convolutional Neural Network, Hacker Intrusion Detection, Support Vector Machine, Text Classification, Word2vec",

author = "Tongbo Wang and Kaoru Hirota and Chaojun Wang and Dongyun Kim 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, T, Hirota, K, Wang, C, Kim, D & Dai, Y 2018, 'Hacker intrusion detection based on text classification with word2vec model and convolutional neural network', Paper presented at 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, China, 2/11/18 - 6/11/18.

Hacker intrusion detection based on text classification with word2vec model and convolutional neural network. / Wang, Tongbo; Hirota, Kaoru; Wang, Chaojun et al.
2018. Paper presented at 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, China.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Hacker intrusion detection based on text classification with word2vec model and convolutional neural network

AU - Wang, Tongbo

AU - Hirota, Kaoru

AU - Wang, Chaojun

AU - Kim, Dongyun

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 hacker intrusion detection method is proposed to tighten up Windows system security with text classification, which the Word2vec model and Convolutional Neural Network (CNN) are applied to distinguish hostile intrusion logs from a benign. The proposal makes full use of users’ operation commands contained in Windows system logs to analyze possible intrusion information and it supplies an application of text classification with CNN and word2vec on network security. For experiment, the dataset from a company having thousands of employees is converted to word vectors by word2vec model. The results of CNN with word embedding and classifier Support Vector Machine (SVM) with term frequency-inverse document frequency (TF-IDF) show the intrusion detection rates 95.56% and 87.55%, respectively. The proposed method can be integrated into antivirus software for hacker intrusion detection.

AB - A hacker intrusion detection method is proposed to tighten up Windows system security with text classification, which the Word2vec model and Convolutional Neural Network (CNN) are applied to distinguish hostile intrusion logs from a benign. The proposal makes full use of users’ operation commands contained in Windows system logs to analyze possible intrusion information and it supplies an application of text classification with CNN and word2vec on network security. For experiment, the dataset from a company having thousands of employees is converted to word vectors by word2vec model. The results of CNN with word embedding and classifier Support Vector Machine (SVM) with term frequency-inverse document frequency (TF-IDF) show the intrusion detection rates 95.56% and 87.55%, respectively. The proposed method can be integrated into antivirus software for hacker intrusion detection.

KW - Convolutional Neural Network

KW - Hacker Intrusion Detection

KW - Support Vector Machine

KW - Text Classification

KW - Word2vec

UR - https://www.scopus.com/pages/publications/85057964568

M3 - Paper

AN - SCOPUS:85057964568

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 -

Wang T, Hirota K, Wang C, Kim D, Dai Y. Hacker intrusion detection based on text classification with word2vec model and convolutional neural network. 2018. Paper presented at 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, China.

Hacker intrusion detection based on text classification with word2vec model and convolutional neural network

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this