Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks

Yusuke Tokuyama; Ryo Miki; Yukinobu Fukushima; Yuya Tarutani; Tokumi Yokohira

doi:10.1145/3395245.3396441

Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks

Yusuke Tokuyama, Ryo Miki, Yukinobu Fukushima, Yuya Tarutani, Tokumi Yokohira

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

3 Citations (Scopus)

Abstract

Recurrent neural network method considering traffic volume, timestamp and day of the week (RNN-VTD method) is a promising network traffic prediction method because of its high prediction accuracy. The RNN-VTD method encodes timestamp and day of the week, which are categorical data, to numerical data using label encoding. The label encoding, however, gives magnitude to the encoded values, which may cause misunderstanding of recurrent neural network models, and consequently, the prediction accuracy of the RNN-VTD method may be degraded. In this paper, we investigate the effect of using one-hot encoding instead of label encoding for a feature encoding method in the RNN-VTD method. In the one-hot encoding, each input data is encoded to an k-dimensional 0 - 1 vector where k is the number of category types. Because the encoded data do not have magnitude, it is expected that the prediction accuracy of the RNN-VTD method is improved.

Original language	English
Title of host publication	Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020
Publisher	Association for Computing Machinery
Pages	279-283
Number of pages	5
ISBN (Electronic)	9781450377058
DOIs	https://doi.org/10.1145/3395245.3396441
Publication status	Published - Mar 28 2020
Event	8th International Conference on Information and Education Technology, ICIET 2020 - Okayama, Japan Duration: Mar 28 2020 → Mar 30 2020

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	8th International Conference on Information and Education Technology, ICIET 2020
Country/Territory	Japan
City	Okayama
Period	3/28/20 → 3/30/20

Keywords

Feature Encoding
Network traffic prediction
Recurrent Neural Networks

ASJC Scopus subject areas

Software
Human-Computer Interaction
Computer Vision and Pattern Recognition
Computer Networks and Communications

Access to Document

10.1145/3395245.3396441

Cite this

Tokuyama, Y., Miki, R., Fukushima, Y., Tarutani, Y., & Yokohira, T. (2020). Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks. In Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020 (pp. 279-283). (ACM International Conference Proceeding Series). Association for Computing Machinery. https://doi.org/10.1145/3395245.3396441

Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks. / Tokuyama, Yusuke; Miki, Ryo; Fukushima, Yukinobu et al.
Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020. Association for Computing Machinery, 2020. p. 279-283 (ACM International Conference Proceeding Series).

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

Tokuyama, Y, Miki, R, Fukushima, Y , Tarutani, Y & Yokohira, T 2020, Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks. in Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 279-283, 8th International Conference on Information and Education Technology, ICIET 2020, Okayama, Japan, 3/28/20. https://doi.org/10.1145/3395245.3396441

Tokuyama Y, Miki R, Fukushima Y , Tarutani Y , Yokohira T. Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks. In Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020. Association for Computing Machinery. 2020. p. 279-283. (ACM International Conference Proceeding Series). doi: 10.1145/3395245.3396441

Tokuyama, Yusuke ; Miki, Ryo ; Fukushima, Yukinobu et al. / Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks. Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020. Association for Computing Machinery, 2020. pp. 279-283 (ACM International Conference Proceeding Series).

@inproceedings{614bef154192452c9de859cfb4498cff,

title = "Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks",

abstract = "Recurrent neural network method considering traffic volume, timestamp and day of the week (RNN-VTD method) is a promising network traffic prediction method because of its high prediction accuracy. The RNN-VTD method encodes timestamp and day of the week, which are categorical data, to numerical data using label encoding. The label encoding, however, gives magnitude to the encoded values, which may cause misunderstanding of recurrent neural network models, and consequently, the prediction accuracy of the RNN-VTD method may be degraded. In this paper, we investigate the effect of using one-hot encoding instead of label encoding for a feature encoding method in the RNN-VTD method. In the one-hot encoding, each input data is encoded to an k-dimensional 0 - 1 vector where k is the number of category types. Because the encoded data do not have magnitude, it is expected that the prediction accuracy of the RNN-VTD method is improved.",

keywords = "Feature Encoding, Network traffic prediction, Recurrent Neural Networks",

author = "Yusuke Tokuyama and Ryo Miki and Yukinobu Fukushima and Yuya Tarutani and Tokumi Yokohira",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.; 8th International Conference on Information and Education Technology, ICIET 2020 ; Conference date: 28-03-2020 Through 30-03-2020",

year = "2020",

month = mar,

day = "28",

doi = "10.1145/3395245.3396441",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "279--283",

booktitle = "Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020",

}

TY - GEN

T1 - Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks

AU - Tokuyama, Yusuke

AU - Miki, Ryo

AU - Fukushima, Yukinobu

AU - Tarutani, Yuya

AU - Yokohira, Tokumi

PY - 2020/3/28

Y1 - 2020/3/28

N2 - Recurrent neural network method considering traffic volume, timestamp and day of the week (RNN-VTD method) is a promising network traffic prediction method because of its high prediction accuracy. The RNN-VTD method encodes timestamp and day of the week, which are categorical data, to numerical data using label encoding. The label encoding, however, gives magnitude to the encoded values, which may cause misunderstanding of recurrent neural network models, and consequently, the prediction accuracy of the RNN-VTD method may be degraded. In this paper, we investigate the effect of using one-hot encoding instead of label encoding for a feature encoding method in the RNN-VTD method. In the one-hot encoding, each input data is encoded to an k-dimensional 0 - 1 vector where k is the number of category types. Because the encoded data do not have magnitude, it is expected that the prediction accuracy of the RNN-VTD method is improved.

AB - Recurrent neural network method considering traffic volume, timestamp and day of the week (RNN-VTD method) is a promising network traffic prediction method because of its high prediction accuracy. The RNN-VTD method encodes timestamp and day of the week, which are categorical data, to numerical data using label encoding. The label encoding, however, gives magnitude to the encoded values, which may cause misunderstanding of recurrent neural network models, and consequently, the prediction accuracy of the RNN-VTD method may be degraded. In this paper, we investigate the effect of using one-hot encoding instead of label encoding for a feature encoding method in the RNN-VTD method. In the one-hot encoding, each input data is encoded to an k-dimensional 0 - 1 vector where k is the number of category types. Because the encoded data do not have magnitude, it is expected that the prediction accuracy of the RNN-VTD method is improved.

KW - Feature Encoding

KW - Network traffic prediction

KW - Recurrent Neural Networks

UR - http://www.scopus.com/inward/record.url?scp=85085944435&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85085944435&partnerID=8YFLogxK

U2 - 10.1145/3395245.3396441

DO - 10.1145/3395245.3396441

M3 - Conference contribution

AN - SCOPUS:85085944435

T3 - ACM International Conference Proceeding Series

SP - 279

EP - 283

BT - Proceedings of the 2020 8th International Conference on Information and Education Technology, ICIET 2020

PB - Association for Computing Machinery

T2 - 8th International Conference on Information and Education Technology, ICIET 2020

Y2 - 28 March 2020 through 30 March 2020

ER -

Performance Evaluation of Feature Encoding Methods in Network Traffic Prediction Using Recurrent Neural Networks

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this