A comparative study of social network classifiers for predicting churn in the telecommunication industry

Maria Oskarsdottir; Cristian Bravo; Wouter Verbeke; Carlos Sarraute; Bart Baesens; Jan Vanthienen

doi:10.1109/ASONAM.2016.7752384

A comparative study of social network classifiers for predicting churn in the telecommunication industry

Maria Oskarsdottir, Cristian Bravo, Wouter Verbeke, Carlos Sarraute, Bart Baesens, Jan Vanthienen

Department of Computer Science

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

16 Citations (Scopus)

Abstract

Relational learning in networked data has been shown to be effective in a number of studies. Relational learners, composed of relational classifiers and collective inference methods, enable the inference of nodes in a network given the existence and strength of links to other nodes. These methods have been adapted to predict customer churn in telecommunication companies showing that incorporating them may give more accurate predictions. In this research, the performance of a variety of relational learners is compared by applying them to a number of CDR datasets originating from the telecommunication industry, with the goal to rank them as a whole and investigate the effects of relational classifiers and collective inference methods separately. Our results show that collective inference methods do not improve the performance of relational classifiers and the best performing relational classifier is the network-only link-based classifier, which builds a logistic model using link-based measures for the nodes in the network.

Original language	English
Title of host publication	Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
Editors	Ravi Kumar, James Caverlee, Hanghang Tong
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1151-1158
Number of pages	8
ISBN (Electronic)	9781509028467
DOIs	https://doi.org/10.1109/ASONAM.2016.7752384
Publication status	Published - 21 Nov 2016
Event	2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016 - San Francisco, United States Duration: 18 Aug 2016 → 21 Aug 2016

Publication series

Name	Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

Conference

Conference	2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
Country/Territory	United States
City	San Francisco
Period	18/08/16 → 21/08/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Access to Document

10.1109/ASONAM.2016.7752384

Cite this

Oskarsdottir, M., Bravo, C., Verbeke, W., Sarraute, C., Baesens, B., & Vanthienen, J. (2016). A comparative study of social network classifiers for predicting churn in the telecommunication industry. In R. Kumar, J. Caverlee, & H. Tong (Eds.), Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016 (pp. 1151-1158). Article 7752384 (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASONAM.2016.7752384

Oskarsdottir, Maria ; Bravo, Cristian ; Verbeke, Wouter et al. / A comparative study of social network classifiers for predicting churn in the telecommunication industry. Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016. editor / Ravi Kumar ; James Caverlee ; Hanghang Tong. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 1151-1158 (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016).

@inproceedings{21ed6fedcdc54b2881510a0bf56f3fc9,

title = "A comparative study of social network classifiers for predicting churn in the telecommunication industry",

abstract = "Relational learning in networked data has been shown to be effective in a number of studies. Relational learners, composed of relational classifiers and collective inference methods, enable the inference of nodes in a network given the existence and strength of links to other nodes. These methods have been adapted to predict customer churn in telecommunication companies showing that incorporating them may give more accurate predictions. In this research, the performance of a variety of relational learners is compared by applying them to a number of CDR datasets originating from the telecommunication industry, with the goal to rank them as a whole and investigate the effects of relational classifiers and collective inference methods separately. Our results show that collective inference methods do not improve the performance of relational classifiers and the best performing relational classifier is the network-only link-based classifier, which builds a logistic model using link-based measures for the nodes in the network.",

author = "Maria Oskarsdottir and Cristian Bravo and Wouter Verbeke and Carlos Sarraute and Bart Baesens and Jan Vanthienen",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016 ; Conference date: 18-08-2016 Through 21-08-2016",

year = "2016",

month = nov,

day = "21",

doi = "10.1109/ASONAM.2016.7752384",

language = "English",

series = "Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016",

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

pages = "1151--1158",

editor = "Ravi Kumar and James Caverlee and Hanghang Tong",

booktitle = "Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016",

address = "United States",

}

Oskarsdottir, M, Bravo, C, Verbeke, W, Sarraute, C, Baesens, B & Vanthienen, J 2016, A comparative study of social network classifiers for predicting churn in the telecommunication industry. in R Kumar, J Caverlee & H Tong (eds), Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016., 7752384, Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016, Institute of Electrical and Electronics Engineers Inc., pp. 1151-1158, 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016, San Francisco, United States, 18/08/16. https://doi.org/10.1109/ASONAM.2016.7752384

A comparative study of social network classifiers for predicting churn in the telecommunication industry. / Oskarsdottir, Maria; Bravo, Cristian; Verbeke, Wouter et al.
Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016. ed. / Ravi Kumar; James Caverlee; Hanghang Tong. Institute of Electrical and Electronics Engineers Inc., 2016. p. 1151-1158 7752384 (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016).

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

TY - GEN

T1 - A comparative study of social network classifiers for predicting churn in the telecommunication industry

AU - Oskarsdottir, Maria

AU - Bravo, Cristian

AU - Verbeke, Wouter

AU - Sarraute, Carlos

AU - Baesens, Bart

AU - Vanthienen, Jan

PY - 2016/11/21

Y1 - 2016/11/21

N2 - Relational learning in networked data has been shown to be effective in a number of studies. Relational learners, composed of relational classifiers and collective inference methods, enable the inference of nodes in a network given the existence and strength of links to other nodes. These methods have been adapted to predict customer churn in telecommunication companies showing that incorporating them may give more accurate predictions. In this research, the performance of a variety of relational learners is compared by applying them to a number of CDR datasets originating from the telecommunication industry, with the goal to rank them as a whole and investigate the effects of relational classifiers and collective inference methods separately. Our results show that collective inference methods do not improve the performance of relational classifiers and the best performing relational classifier is the network-only link-based classifier, which builds a logistic model using link-based measures for the nodes in the network.

AB - Relational learning in networked data has been shown to be effective in a number of studies. Relational learners, composed of relational classifiers and collective inference methods, enable the inference of nodes in a network given the existence and strength of links to other nodes. These methods have been adapted to predict customer churn in telecommunication companies showing that incorporating them may give more accurate predictions. In this research, the performance of a variety of relational learners is compared by applying them to a number of CDR datasets originating from the telecommunication industry, with the goal to rank them as a whole and investigate the effects of relational classifiers and collective inference methods separately. Our results show that collective inference methods do not improve the performance of relational classifiers and the best performing relational classifier is the network-only link-based classifier, which builds a logistic model using link-based measures for the nodes in the network.

UR - http://www.scopus.com/inward/record.url?scp=85006822885&partnerID=8YFLogxK

U2 - 10.1109/ASONAM.2016.7752384

DO - 10.1109/ASONAM.2016.7752384

M3 - Conference contribution

AN - SCOPUS:85006822885

T3 - Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

SP - 1151

EP - 1158

BT - Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

A2 - Kumar, Ravi

A2 - Caverlee, James

A2 - Tong, Hanghang

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

Y2 - 18 August 2016 through 21 August 2016

ER -

Oskarsdottir M, Bravo C, Verbeke W, Sarraute C, Baesens B, Vanthienen J. A comparative study of social network classifiers for predicting churn in the telecommunication industry. In Kumar R, Caverlee J, Tong H, editors, Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 1151-1158. 7752384. (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016). doi: 10.1109/ASONAM.2016.7752384

A comparative study of social network classifiers for predicting churn in the telecommunication industry

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this