Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification

Gendo Kumoi; Masayuki Goto

doi:10.1109/BCD.2019.8885219

Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification

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

Abstract

In recent years, toward the realization of the Internet of Things ('IoT') society, related technologies have been developed and various electronic devices are being connected to the network. Even in companies that provide such kinds of products and services, it is possible to collect usage histories of their customers. If the companies can appropriately analyze the usage history data, it is useful for their marketing activities. However, in general, device usage histories are multivariate time-series data, and it is not obvious how to construct a feature space for customer classification and clustering. Therefore, this paper proposes a method to automatically select feature quantities characterizing the properties of customers using machine learning. We apply this method to real data and show its effectiveness.

Original language	English
Title of host publication	Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019
Editors	Motoi Iwashita, Atsushi Shimoda, Prajak Chertchom
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	154-159
Number of pages	6
ISBN (Electronic)	9781728108865
DOIs	https://doi.org/10.1109/BCD.2019.8885219
Publication status	Published - 2019 May
Event	4th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019 - Honolulu, United States Duration: 2019 May 29 → 2019 May 31

Publication series

Name	Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019

Conference

Conference	4th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019
Country/Territory	United States
City	Honolulu
Period	19/5/29 → 19/5/31

Keywords

Binary Classifier
Feature Quantity Selection
Multivariate Time Series Data
Random Forest

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Computer Science Applications
Information Systems and Management

Access to Document

10.1109/BCD.2019.8885219

Cite this

Kumoi, G., & Goto, M. (2019). Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification. In M. Iwashita, A. Shimoda, & P. Chertchom (Eds.), Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019 (pp. 154-159). [8885219] (Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BCD.2019.8885219

Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification. / Kumoi, Gendo; Goto, Masayuki.
Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019. ed. / Motoi Iwashita; Atsushi Shimoda; Prajak Chertchom. Institute of Electrical and Electronics Engineers Inc., 2019. p. 154-159 8885219 (Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019).

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

Kumoi, G & Goto, M 2019, Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification. in M Iwashita, A Shimoda & P Chertchom (eds), Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019., 8885219, Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019, Institute of Electrical and Electronics Engineers Inc., pp. 154-159, 4th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019, Honolulu, United States, 19/5/29. https://doi.org/10.1109/BCD.2019.8885219

Kumoi G, Goto M. Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification. In Iwashita M, Shimoda A, Chertchom P, editors, Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 154-159. 8885219. (Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019). doi: 10.1109/BCD.2019.8885219

Kumoi, Gendo ; Goto, Masayuki. / Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification. Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019. editor / Motoi Iwashita ; Atsushi Shimoda ; Prajak Chertchom. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 154-159 (Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019).

@inproceedings{1294d824c75a42a0b81b2d525494b241,

title = "Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification",

abstract = "In recent years, toward the realization of the Internet of Things ('IoT') society, related technologies have been developed and various electronic devices are being connected to the network. Even in companies that provide such kinds of products and services, it is possible to collect usage histories of their customers. If the companies can appropriately analyze the usage history data, it is useful for their marketing activities. However, in general, device usage histories are multivariate time-series data, and it is not obvious how to construct a feature space for customer classification and clustering. Therefore, this paper proposes a method to automatically select feature quantities characterizing the properties of customers using machine learning. We apply this method to real data and show its effectiveness.",

keywords = "Binary Classifier, Feature Quantity Selection, Multivariate Time Series Data, Random Forest",

author = "Gendo Kumoi and Masayuki Goto",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 4th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019 ; Conference date: 29-05-2019 Through 31-05-2019",

year = "2019",

month = may,

doi = "10.1109/BCD.2019.8885219",

language = "English",

series = "Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "154--159",

editor = "Motoi Iwashita and Atsushi Shimoda and Prajak Chertchom",

booktitle = "Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019",

}

TY - GEN

T1 - Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification

AU - Kumoi, Gendo

AU - Goto, Masayuki

PY - 2019/5

Y1 - 2019/5

N2 - In recent years, toward the realization of the Internet of Things ('IoT') society, related technologies have been developed and various electronic devices are being connected to the network. Even in companies that provide such kinds of products and services, it is possible to collect usage histories of their customers. If the companies can appropriately analyze the usage history data, it is useful for their marketing activities. However, in general, device usage histories are multivariate time-series data, and it is not obvious how to construct a feature space for customer classification and clustering. Therefore, this paper proposes a method to automatically select feature quantities characterizing the properties of customers using machine learning. We apply this method to real data and show its effectiveness.

AB - In recent years, toward the realization of the Internet of Things ('IoT') society, related technologies have been developed and various electronic devices are being connected to the network. Even in companies that provide such kinds of products and services, it is possible to collect usage histories of their customers. If the companies can appropriately analyze the usage history data, it is useful for their marketing activities. However, in general, device usage histories are multivariate time-series data, and it is not obvious how to construct a feature space for customer classification and clustering. Therefore, this paper proposes a method to automatically select feature quantities characterizing the properties of customers using machine learning. We apply this method to real data and show its effectiveness.

KW - Binary Classifier

KW - Feature Quantity Selection

KW - Multivariate Time Series Data

KW - Random Forest

UR - http://www.scopus.com/inward/record.url?scp=85075025856&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075025856&partnerID=8YFLogxK

U2 - 10.1109/BCD.2019.8885219

DO - 10.1109/BCD.2019.8885219

M3 - Conference contribution

AN - SCOPUS:85075025856

T3 - Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019

SP - 154

EP - 159

BT - Proceedings - 2019 IEEE/ACIS 4th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019

A2 - Iwashita, Motoi

A2 - Shimoda, Atsushi

A2 - Chertchom, Prajak

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 4th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2019

Y2 - 29 May 2019 through 31 May 2019

ER -

Greedy Features Quantity Selection Method from Multivariate Time Series Data for Customer Classification

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this