Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection

Chenmao Lu; Hong Ning Dai; Junhao Zhou; Hao Wang

doi:10.1007/978-3-030-69992-5_5

Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection

Chenmao Lu, Hong Ning Dai^*, Junhao Zhou, Hao Wang

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

Cloud computing offers elastic and ubiquitous computing services, thereby receiving extensive attention recently. However, cloud servers have also become the targets of malicious attacks or hackers due to the centralization of data storage and computing facilities. Most intrusion attacks to cloud servers are often originated from inner or external networks. Intrusion detection is a prerequisite to designing anti-intrusion countermeasures of cloud systems. In this paper, we explore deep learning algorithms to design intrusion detection methods. In particular, we present a deep learning-based method with the integration of conventional neural networks, self-attention mechanism, and Long short-term memory (LSTM), namely CNN-A-LSTM to detect intrusion. CNN-A-LSTM leverages the merits of CNN in processing local correlation data and extracting features, the time feature extracting capability of LSTM, and the self-attention mechanism to better exact features. We conduct extensive experiments on the KDDcup99 dataset to evaluate the performance of our CNN-A-LSTM model. Compared with other machine learning and deep learning models, our CNN-A-LSTM has superior performance.

Original language	English
Title of host publication	Cloud Computing
Subtitle of host publication	10th EAI International Conference, CloudComp 2020, Qufu, China, December 11-12, 2020, Proceedings
Editors	Lianyong Qi, Mohammad R. Khosravi, Xiaolong Xu, Yiwen Zhang, Varun G. Menon
Publisher	Springer Cham
Pages	57-73
Number of pages	17
Edition	1st
ISBN (Electronic)	9783030699925
ISBN (Print)	9783030699918
DOIs	https://doi.org/10.1007/978-3-030-69992-5_5
Publication status	Published - 13 Feb 2021
Event	10th EAI International Conference on Cloud Computing, CloudComp 2020 - Virtual, Online, Qufu, China Duration: 11 Dec 2020 → 12 Dec 2020 https://link.springer.com/book/10.1007/978-3-030-69992-5

Publication series

Name	Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST
Volume	363
ISSN (Print)	1867-8211
ISSN (Electronic)	1867-822X

Name	CloudComp: International Conference on Cloud Computing

Conference

Conference	10th EAI International Conference on Cloud Computing, CloudComp 2020
Country/Territory	China
City	Qufu
Period	11/12/20 → 12/12/20
Internet address	https://link.springer.com/book/10.1007/978-3-030-69992-5

Scopus Subject Areas

Computer Networks and Communications

User-Defined Keywords

Convolution neural network
Deep learning
Long short-term memory
Network intrusion detection
Self-attention

Access to Document

10.1007/978-3-030-69992-5_5

Cite this

Lu, C., Dai, H. N., Zhou, J., & Wang, H. (2021). Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection. In L. Qi, M. R. Khosravi, X. Xu, Y. Zhang, & V. G. Menon (Eds.), Cloud Computing: 10th EAI International Conference, CloudComp 2020, Qufu, China, December 11-12, 2020, Proceedings (1st ed., pp. 57-73). (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST; Vol. 363), (CloudComp: International Conference on Cloud Computing). Springer Cham. https://doi.org/10.1007/978-3-030-69992-5_5

Lu, Chenmao ; Dai, Hong Ning ; Zhou, Junhao et al. / Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection. Cloud Computing: 10th EAI International Conference, CloudComp 2020, Qufu, China, December 11-12, 2020, Proceedings. editor / Lianyong Qi ; Mohammad R. Khosravi ; Xiaolong Xu ; Yiwen Zhang ; Varun G. Menon. 1st. ed. Springer Cham, 2021. pp. 57-73 (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST). (CloudComp: International Conference on Cloud Computing).

@inproceedings{39a14c3f66564c5a8c89f29e98f0f719,

title = "Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection",

abstract = "Cloud computing offers elastic and ubiquitous computing services, thereby receiving extensive attention recently. However, cloud servers have also become the targets of malicious attacks or hackers due to the centralization of data storage and computing facilities. Most intrusion attacks to cloud servers are often originated from inner or external networks. Intrusion detection is a prerequisite to designing anti-intrusion countermeasures of cloud systems. In this paper, we explore deep learning algorithms to design intrusion detection methods. In particular, we present a deep learning-based method with the integration of conventional neural networks, self-attention mechanism, and Long short-term memory (LSTM), namely CNN-A-LSTM to detect intrusion. CNN-A-LSTM leverages the merits of CNN in processing local correlation data and extracting features, the time feature extracting capability of LSTM, and the self-attention mechanism to better exact features. We conduct extensive experiments on the KDDcup99 dataset to evaluate the performance of our CNN-A-LSTM model. Compared with other machine learning and deep learning models, our CNN-A-LSTM has superior performance.",

keywords = "Convolution neural network, Deep learning, Long short-term memory, Network intrusion detection, Self-attention",

author = "Chenmao Lu and Dai, {Hong Ning} and Junhao Zhou and Hao Wang",

note = "Funding Information: Acknowledgement. The work described in this paper was partially supported by Macao Science and Technology Development Fund under Grant No. 0026/2018/A1. Publisher Copyright: {\textcopyright} 2021, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.; 10th EAI International Conference on Cloud Computing, CloudComp 2020 ; Conference date: 11-12-2020 Through 12-12-2020",

year = "2021",

month = feb,

day = "13",

doi = "10.1007/978-3-030-69992-5_5",

language = "English",

isbn = "9783030699918",

series = "Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST",

publisher = "Springer Cham",

pages = "57--73",

editor = "Lianyong Qi and Khosravi, {Mohammad R.} and Xiaolong Xu and Yiwen Zhang and Menon, {Varun G.}",

booktitle = "Cloud Computing",

edition = "1st",

url = "https://link.springer.com/book/10.1007/978-3-030-69992-5",

}

Lu, C, Dai, HN, Zhou, J & Wang, H 2021, Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection. in L Qi, MR Khosravi, X Xu, Y Zhang & VG Menon (eds), Cloud Computing: 10th EAI International Conference, CloudComp 2020, Qufu, China, December 11-12, 2020, Proceedings. 1st edn, Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST, vol. 363, CloudComp: International Conference on Cloud Computing, Springer Cham, pp. 57-73, 10th EAI International Conference on Cloud Computing, CloudComp 2020, Qufu, China, 11/12/20. https://doi.org/10.1007/978-3-030-69992-5_5

Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection. / Lu, Chenmao; Dai, Hong Ning; Zhou, Junhao et al.

Cloud Computing: 10th EAI International Conference, CloudComp 2020, Qufu, China, December 11-12, 2020, Proceedings. ed. / Lianyong Qi; Mohammad R. Khosravi; Xiaolong Xu; Yiwen Zhang; Varun G. Menon. 1st. ed. Springer Cham, 2021. p. 57-73 (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST; Vol. 363), (CloudComp: International Conference on Cloud Computing).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection

AU - Lu, Chenmao

AU - Dai, Hong Ning

AU - Zhou, Junhao

AU - Wang, Hao

N1 - Funding Information: Acknowledgement. The work described in this paper was partially supported by Macao Science and Technology Development Fund under Grant No. 0026/2018/A1. Publisher Copyright: © 2021, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

PY - 2021/2/13

Y1 - 2021/2/13

N2 - Cloud computing offers elastic and ubiquitous computing services, thereby receiving extensive attention recently. However, cloud servers have also become the targets of malicious attacks or hackers due to the centralization of data storage and computing facilities. Most intrusion attacks to cloud servers are often originated from inner or external networks. Intrusion detection is a prerequisite to designing anti-intrusion countermeasures of cloud systems. In this paper, we explore deep learning algorithms to design intrusion detection methods. In particular, we present a deep learning-based method with the integration of conventional neural networks, self-attention mechanism, and Long short-term memory (LSTM), namely CNN-A-LSTM to detect intrusion. CNN-A-LSTM leverages the merits of CNN in processing local correlation data and extracting features, the time feature extracting capability of LSTM, and the self-attention mechanism to better exact features. We conduct extensive experiments on the KDDcup99 dataset to evaluate the performance of our CNN-A-LSTM model. Compared with other machine learning and deep learning models, our CNN-A-LSTM has superior performance.

AB - Cloud computing offers elastic and ubiquitous computing services, thereby receiving extensive attention recently. However, cloud servers have also become the targets of malicious attacks or hackers due to the centralization of data storage and computing facilities. Most intrusion attacks to cloud servers are often originated from inner or external networks. Intrusion detection is a prerequisite to designing anti-intrusion countermeasures of cloud systems. In this paper, we explore deep learning algorithms to design intrusion detection methods. In particular, we present a deep learning-based method with the integration of conventional neural networks, self-attention mechanism, and Long short-term memory (LSTM), namely CNN-A-LSTM to detect intrusion. CNN-A-LSTM leverages the merits of CNN in processing local correlation data and extracting features, the time feature extracting capability of LSTM, and the self-attention mechanism to better exact features. We conduct extensive experiments on the KDDcup99 dataset to evaluate the performance of our CNN-A-LSTM model. Compared with other machine learning and deep learning models, our CNN-A-LSTM has superior performance.

KW - Convolution neural network

KW - Deep learning

KW - Long short-term memory

KW - Network intrusion detection

KW - Self-attention

UR - http://www.scopus.com/inward/record.url?scp=85102654028&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-69992-5_5

DO - 10.1007/978-3-030-69992-5_5

M3 - Conference proceeding

AN - SCOPUS:85102654028

SN - 9783030699918

T3 - Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST

SP - 57

EP - 73

BT - Cloud Computing

A2 - Qi, Lianyong

A2 - Khosravi, Mohammad R.

A2 - Xu, Xiaolong

A2 - Zhang, Yiwen

A2 - Menon, Varun G.

PB - Springer Cham

T2 - 10th EAI International Conference on Cloud Computing, CloudComp 2020

Y2 - 11 December 2020 through 12 December 2020

ER -

Lu C, Dai HN, Zhou J, Wang H. Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection. In Qi L, Khosravi MR, Xu X, Zhang Y, Menon VG, editors, Cloud Computing: 10th EAI International Conference, CloudComp 2020, Qufu, China, December 11-12, 2020, Proceedings. 1st ed. Springer Cham. 2021. p. 57-73. (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST). (CloudComp: International Conference on Cloud Computing). doi: 10.1007/978-3-030-69992-5_5

Exploring Self-attention Mechanism of Deep Learning in Cloud Intrusion Detection

Abstract

Publication series

Conference

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this