Prediction of bank telemarketing with co-training of mixture-of-experts and MLP

Jae Min Yu, Sung-Bae Cho

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Utilization of financial data becomes one of the important issues for user adaptive marketing on the bank service. The marketing is conducted based on personal information with various facts that affect a success (clients agree to accept financial instrument). Personal information can be collected continuously anytime if clients want to agree to use own information in case of opening an account in bank, but labeling all the data needs to pay a high cost. In this paper, focusing on this characteristics of financial data, we present a global-local co-training (GLCT) algorithm to utilize labeled and unlabeled data to construct better prediction model. We performed experiments using real-world data from Portuguese bank. Experiments show that GLCT performs well regardless of the ratio of initial labeled data. Through the series of iterating experiments, we obtained better results on various aspects.

Original languageEnglish
Title of host publicationNeural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
EditorsKazushi Ikeda, Minho Lee, Akira Hirose, Seiichi Ozawa, Kenji Doya, Derong Liu
PublisherSpringer Verlag
Pages52-59
Number of pages8
ISBN (Print)9783319466804
DOIs
Publication statusPublished - 2016 Jan 1
Event23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, Japan
Duration: 2016 Oct 162016 Oct 21

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9950 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other23rd International Conference on Neural Information Processing, ICONIP 2016
CountryJapan
CityKyoto
Period16/10/1616/10/21

Fingerprint

Co-training
Mixture of Experts
Financial Data
Marketing
Prediction
Experiments
Information use
Experiment
Labeling
Training Algorithm
Prediction Model
Series
Banks
Costs

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Yu, J. M., & Cho, S-B. (2016). Prediction of bank telemarketing with co-training of mixture-of-experts and MLP. In K. Ikeda, M. Lee, A. Hirose, S. Ozawa, K. Doya, & D. Liu (Eds.), Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings (pp. 52-59). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9950 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-46681-1_7
Yu, Jae Min ; Cho, Sung-Bae. / Prediction of bank telemarketing with co-training of mixture-of-experts and MLP. Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. editor / Kazushi Ikeda ; Minho Lee ; Akira Hirose ; Seiichi Ozawa ; Kenji Doya ; Derong Liu. Springer Verlag, 2016. pp. 52-59 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{e59ee0bd1d2b47ebb93828fda13d49c7,
title = "Prediction of bank telemarketing with co-training of mixture-of-experts and MLP",
abstract = "Utilization of financial data becomes one of the important issues for user adaptive marketing on the bank service. The marketing is conducted based on personal information with various facts that affect a success (clients agree to accept financial instrument). Personal information can be collected continuously anytime if clients want to agree to use own information in case of opening an account in bank, but labeling all the data needs to pay a high cost. In this paper, focusing on this characteristics of financial data, we present a global-local co-training (GLCT) algorithm to utilize labeled and unlabeled data to construct better prediction model. We performed experiments using real-world data from Portuguese bank. Experiments show that GLCT performs well regardless of the ratio of initial labeled data. Through the series of iterating experiments, we obtained better results on various aspects.",
author = "Yu, {Jae Min} and Sung-Bae Cho",
year = "2016",
month = "1",
day = "1",
doi = "10.1007/978-3-319-46681-1_7",
language = "English",
isbn = "9783319466804",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "52--59",
editor = "Kazushi Ikeda and Minho Lee and Akira Hirose and Seiichi Ozawa and Kenji Doya and Derong Liu",
booktitle = "Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings",
address = "Germany",

}

Yu, JM & Cho, S-B 2016, Prediction of bank telemarketing with co-training of mixture-of-experts and MLP. in K Ikeda, M Lee, A Hirose, S Ozawa, K Doya & D Liu (eds), Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9950 LNCS, Springer Verlag, pp. 52-59, 23rd International Conference on Neural Information Processing, ICONIP 2016, Kyoto, Japan, 16/10/16. https://doi.org/10.1007/978-3-319-46681-1_7

Prediction of bank telemarketing with co-training of mixture-of-experts and MLP. / Yu, Jae Min; Cho, Sung-Bae.

Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. ed. / Kazushi Ikeda; Minho Lee; Akira Hirose; Seiichi Ozawa; Kenji Doya; Derong Liu. Springer Verlag, 2016. p. 52-59 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9950 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Prediction of bank telemarketing with co-training of mixture-of-experts and MLP

AU - Yu, Jae Min

AU - Cho, Sung-Bae

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Utilization of financial data becomes one of the important issues for user adaptive marketing on the bank service. The marketing is conducted based on personal information with various facts that affect a success (clients agree to accept financial instrument). Personal information can be collected continuously anytime if clients want to agree to use own information in case of opening an account in bank, but labeling all the data needs to pay a high cost. In this paper, focusing on this characteristics of financial data, we present a global-local co-training (GLCT) algorithm to utilize labeled and unlabeled data to construct better prediction model. We performed experiments using real-world data from Portuguese bank. Experiments show that GLCT performs well regardless of the ratio of initial labeled data. Through the series of iterating experiments, we obtained better results on various aspects.

AB - Utilization of financial data becomes one of the important issues for user adaptive marketing on the bank service. The marketing is conducted based on personal information with various facts that affect a success (clients agree to accept financial instrument). Personal information can be collected continuously anytime if clients want to agree to use own information in case of opening an account in bank, but labeling all the data needs to pay a high cost. In this paper, focusing on this characteristics of financial data, we present a global-local co-training (GLCT) algorithm to utilize labeled and unlabeled data to construct better prediction model. We performed experiments using real-world data from Portuguese bank. Experiments show that GLCT performs well regardless of the ratio of initial labeled data. Through the series of iterating experiments, we obtained better results on various aspects.

UR - http://www.scopus.com/inward/record.url?scp=84992761611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992761611&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-46681-1_7

DO - 10.1007/978-3-319-46681-1_7

M3 - Conference contribution

SN - 9783319466804

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 52

EP - 59

BT - Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings

A2 - Ikeda, Kazushi

A2 - Lee, Minho

A2 - Hirose, Akira

A2 - Ozawa, Seiichi

A2 - Doya, Kenji

A2 - Liu, Derong

PB - Springer Verlag

ER -

Yu JM, Cho S-B. Prediction of bank telemarketing with co-training of mixture-of-experts and MLP. In Ikeda K, Lee M, Hirose A, Ozawa S, Doya K, Liu D, editors, Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Springer Verlag. 2016. p. 52-59. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-46681-1_7