An empirical comparison of techniques for the class imbalance problem in churn prediction
An empirical comparison of techniques for the class imbalance problem in churn prediction
Class imbalance brings significant challenges to customer churn prediction. Many solutions have been developed to address this issue. In this paper, we comprehensively compare the performance of state-of-the-art techniques to deal with class imbalance in the context of churn prediction. A recently developed expected maximum profit criterion is used as one of the main performance measures to offer more insights from the perspective of cost- benefit. The experimental results show that the applied evaluation metric has a great im- pact on the performance of techniques. An in-depth exploration of reaction patterns to dif- ferent measures is conducted by intra-family comparison within each solution group and global comparison among the representative techniques from different groups. The results also indicate there is much space to improve solutions’ performance in terms of profit- based measure. Our study offers valuable insights for academics and professionals and it also provides a baseline to develop new methods for dealing with class imbalance in churn prediction.
84-89
Zhu, B.
ac66f851-8282-4ebf-918e-284693cc1d4f
Baesens, B.
f7c6496b-aa7f-4026-8616-ca61d9e216f0
Van Den Brouckes, S.
509bb02a-0bdc-45c0-9fc8-1d2175fdd351
October 2017
Zhu, B.
ac66f851-8282-4ebf-918e-284693cc1d4f
Baesens, B.
f7c6496b-aa7f-4026-8616-ca61d9e216f0
Van Den Brouckes, S.
509bb02a-0bdc-45c0-9fc8-1d2175fdd351
Zhu, B., Baesens, B. and Van Den Brouckes, S.
(2017)
An empirical comparison of techniques for the class imbalance problem in churn prediction.
Information Sciences, 408, .
(doi:10.1016/j.ins.2017.04.015).
Abstract
Class imbalance brings significant challenges to customer churn prediction. Many solutions have been developed to address this issue. In this paper, we comprehensively compare the performance of state-of-the-art techniques to deal with class imbalance in the context of churn prediction. A recently developed expected maximum profit criterion is used as one of the main performance measures to offer more insights from the perspective of cost- benefit. The experimental results show that the applied evaluation metric has a great im- pact on the performance of techniques. An in-depth exploration of reaction patterns to dif- ferent measures is conducted by intra-family comparison within each solution group and global comparison among the representative techniques from different groups. The results also indicate there is much space to improve solutions’ performance in terms of profit- based measure. Our study offers valuable insights for academics and professionals and it also provides a baseline to develop new methods for dealing with class imbalance in churn prediction.
Text
An empirical comparison of techniques for the class imbalance problem in churn prediction
- Accepted Manuscript
Text
An empirical comparison of techniques for the class imbalance problem in churn prediction
- Version of Record
Restricted to Repository staff only
Request a copy
More information
Accepted/In Press date: 11 April 2017
e-pub ahead of print date: 18 April 2017
Published date: October 2017
Identifiers
Local EPrints ID: 413694
URI: http://eprints.soton.ac.uk/id/eprint/413694
ISSN: 0020-0255
PURE UUID: 37ce6948-15bf-4727-bc81-2cf5c6382a42
Catalogue record
Date deposited: 31 Aug 2017 16:31
Last modified: 16 Mar 2024 03:39
Export record
Altmetrics
Contributors
Author:
B. Zhu
Author:
S. Van Den Brouckes
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
