Attention-Based Multi-component LSTM for Internet Traffic Prediction

Qian Xu; Zhenjie Yao; Yanhui Tu; Yixin Chen

doi:10.1007/978-3-030-63823-8_87

Attention-Based Multi-component LSTM for Internet Traffic Prediction

Qian Xu
, Zhenjie Yao
, Yanhui Tu
, Yixin Chen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

With the rapid development of Internet technology, various kinds of electronic products such as mobile phones and laptops become widely available and our daily lives rely more and more on the Internet. Increasing network access brings a series of problems for Internet Service Provider (ISP), such as network congestion and network resource allocation. Effective network traffic prediction can alleviate the aforementioned problems by estimating future traffic based on historical data. In this paper, we propose a novel model named Attention-based Multi-Component LSTM (AMC-LSTM) for Internet traffic prediction. The proposed model is composed of three independent LSTM components, including hour component, day component and week component, to jointly forecast future network traffic with historical data. Moreover, attention mechanism is incorporated into each component to capture the most informative time steps. Experimental results on real-world datasets demonstrate the effectiveness of our model.

Original language	English
Title of host publication	Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings
Editors	Haiqin Yang, Kitsuchart Pasupa, Andrew Chi-Sing Leung, James T. Kwok, Jonathan H. Chan, Irwin King
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	770-777
Number of pages	8
ISBN (Print)	9783030638221
DOIs	https://doi.org/10.1007/978-3-030-63823-8_87
State	Published - 2020
Event	27th International Conference on Neural Information Processing, ICONIP 2020 - Bangkok, Thailand Duration: Nov 18 2020 → Nov 22 2020

Publication series

Name	Communications in Computer and Information Science
Volume	1333
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	27th International Conference on Neural Information Processing, ICONIP 2020
Country/Territory	Thailand
City	Bangkok
Period	11/18/20 → 11/22/20

Keywords

Attention
Internet traffic prediction
LSTM
Multi-component

Access to Document

10.1007/978-3-030-63823-8_87

Cite this

Xu, Q., Yao, Z., Tu, Y., & Chen, Y. (2020). Attention-Based Multi-component LSTM for Internet Traffic Prediction. In H. Yang, K. Pasupa, A. C.-S. Leung, J. T. Kwok, J. H. Chan, & I. King (Eds.), Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings (pp. 770-777). (Communications in Computer and Information Science; Vol. 1333). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-63823-8_87

Xu, Qian ; Yao, Zhenjie ; Tu, Yanhui et al. / Attention-Based Multi-component LSTM for Internet Traffic Prediction. Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings. editor / Haiqin Yang ; Kitsuchart Pasupa ; Andrew Chi-Sing Leung ; James T. Kwok ; Jonathan H. Chan ; Irwin King. Springer Science and Business Media Deutschland GmbH, 2020. pp. 770-777 (Communications in Computer and Information Science).

@inproceedings{0753796eedfd49cb9523e0b43d385067,

title = "Attention-Based Multi-component LSTM for Internet Traffic Prediction",

abstract = "With the rapid development of Internet technology, various kinds of electronic products such as mobile phones and laptops become widely available and our daily lives rely more and more on the Internet. Increasing network access brings a series of problems for Internet Service Provider (ISP), such as network congestion and network resource allocation. Effective network traffic prediction can alleviate the aforementioned problems by estimating future traffic based on historical data. In this paper, we propose a novel model named Attention-based Multi-Component LSTM (AMC-LSTM) for Internet traffic prediction. The proposed model is composed of three independent LSTM components, including hour component, day component and week component, to jointly forecast future network traffic with historical data. Moreover, attention mechanism is incorporated into each component to capture the most informative time steps. Experimental results on real-world datasets demonstrate the effectiveness of our model.",

keywords = "Attention, Internet traffic prediction, LSTM, Multi-component",

author = "Qian Xu and Zhenjie Yao and Yanhui Tu and Yixin Chen",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 27th International Conference on Neural Information Processing, ICONIP 2020 ; Conference date: 18-11-2020 Through 22-11-2020",

year = "2020",

doi = "10.1007/978-3-030-63823-8\_87",

language = "English",

isbn = "9783030638221",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "770--777",

editor = "Haiqin Yang and Kitsuchart Pasupa and Leung, \{Andrew Chi-Sing\} and Kwok, \{James T.\} and Chan, \{Jonathan H.\} and Irwin King",

booktitle = "Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings",

}

Xu, Q, Yao, Z, Tu, Y & Chen, Y 2020, Attention-Based Multi-component LSTM for Internet Traffic Prediction. in H Yang, K Pasupa, AC-S Leung, JT Kwok, JH Chan & I King (eds), Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings. Communications in Computer and Information Science, vol. 1333, Springer Science and Business Media Deutschland GmbH, pp. 770-777, 27th International Conference on Neural Information Processing, ICONIP 2020, Bangkok, Thailand, 11/18/20. https://doi.org/10.1007/978-3-030-63823-8_87

Attention-Based Multi-component LSTM for Internet Traffic Prediction. / Xu, Qian; Yao, Zhenjie; Tu, Yanhui et al.
Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings. ed. / Haiqin Yang; Kitsuchart Pasupa; Andrew Chi-Sing Leung; James T. Kwok; Jonathan H. Chan; Irwin King. Springer Science and Business Media Deutschland GmbH, 2020. p. 770-777 (Communications in Computer and Information Science; Vol. 1333).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Attention-Based Multi-component LSTM for Internet Traffic Prediction

AU - Xu, Qian

AU - Yao, Zhenjie

AU - Tu, Yanhui

AU - Chen, Yixin

PY - 2020

Y1 - 2020

N2 - With the rapid development of Internet technology, various kinds of electronic products such as mobile phones and laptops become widely available and our daily lives rely more and more on the Internet. Increasing network access brings a series of problems for Internet Service Provider (ISP), such as network congestion and network resource allocation. Effective network traffic prediction can alleviate the aforementioned problems by estimating future traffic based on historical data. In this paper, we propose a novel model named Attention-based Multi-Component LSTM (AMC-LSTM) for Internet traffic prediction. The proposed model is composed of three independent LSTM components, including hour component, day component and week component, to jointly forecast future network traffic with historical data. Moreover, attention mechanism is incorporated into each component to capture the most informative time steps. Experimental results on real-world datasets demonstrate the effectiveness of our model.

AB - With the rapid development of Internet technology, various kinds of electronic products such as mobile phones and laptops become widely available and our daily lives rely more and more on the Internet. Increasing network access brings a series of problems for Internet Service Provider (ISP), such as network congestion and network resource allocation. Effective network traffic prediction can alleviate the aforementioned problems by estimating future traffic based on historical data. In this paper, we propose a novel model named Attention-based Multi-Component LSTM (AMC-LSTM) for Internet traffic prediction. The proposed model is composed of three independent LSTM components, including hour component, day component and week component, to jointly forecast future network traffic with historical data. Moreover, attention mechanism is incorporated into each component to capture the most informative time steps. Experimental results on real-world datasets demonstrate the effectiveness of our model.

KW - Attention

KW - Internet traffic prediction

KW - LSTM

KW - Multi-component

UR - https://www.scopus.com/pages/publications/85097095564

U2 - 10.1007/978-3-030-63823-8_87

DO - 10.1007/978-3-030-63823-8_87

M3 - Conference contribution

AN - SCOPUS:85097095564

SN - 9783030638221

T3 - Communications in Computer and Information Science

SP - 770

EP - 777

BT - Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings

A2 - Yang, Haiqin

A2 - Pasupa, Kitsuchart

A2 - Leung, Andrew Chi-Sing

A2 - Kwok, James T.

A2 - Chan, Jonathan H.

A2 - King, Irwin

PB - Springer Science and Business Media Deutschland GmbH

T2 - 27th International Conference on Neural Information Processing, ICONIP 2020

Y2 - 18 November 2020 through 22 November 2020

ER -

Xu Q, Yao Z, Tu Y, Chen Y. Attention-Based Multi-component LSTM for Internet Traffic Prediction. In Yang H, Pasupa K, Leung ACS, Kwok JT, Chan JH, King I, editors, Neural Information Processing - 27th International Conference, ICONIP 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 770-777. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-63823-8_87

Attention-Based Multi-component LSTM for Internet Traffic Prediction

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this