Domain knowledge integration in data mining using decision tables: case studies in churn prediction
Open accessThe DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.
Abstract
Companies' interest in customer relationship modelling and key issues such as customer lifetime value and churn has substantially increased over the years. However, the complexity of building, interpreting and applying these models creates obstacles for their implementation. The main contribution of this paper is to show how domain knowledge can be incorporated in the data mining process for churn prediction, viz. through the evaluation of coefficient signs in a logistic regression model, and secondly, by analysing a decision table (DT) extracted from a decision tree or rule-based classifier. An algorithm to check DTs for violations of monotonicity constraints is presented, which involves the repeated application of condition reordering and table contraction to detect counter-intuitive patterns. Both approaches are applied to two telecom data sets to empirically demonstrate how domain knowledge can be used to ensure the interpretability of the resulting models.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Benchmarking state-of-the-art classification algorithms for credit scoring
Baesens, B., Van Gestel, T., Viaene, S., Stepanova, M., Suykens, J., & Vanthienen, J. (2003) · Journal of the Operational Research Society · doi:10.1057/palgrave.jors.2601545 · Cited by 904 · cited 2×
“The motivation is that both techniques yield simple, easy to understand models and furthermore have been shown to perform well in various problem settings, e.g. for credit scoring (Baesens et al , 2003 b ).”
BackgroundMethodSelf-citationRFM and CLV: Using Iso-Value Curves for Customer Base Analysis
Fader, P. S., Hardie, B. G. S., & Lee, K. L. (2005) · Journal of Marketing Research · doi:10.1509/jmkr.2005.42.4.415 · Cited by 595 · cited 1×
“Customer lifetime value (CLV) (Jain and Singh, 2002; Fader et al , 2005; Malthouse and Blattberg, 2005) represents the present value of the expected benefits minus the costs of initialising, maintaining and developing the customer relationship.”
BackgroundCustomer lifetime value research in marketing: A review and future directions
Jain, D., & Singh, S. S. (2002) · Journal of Interactive Marketing · doi:10.1002/dir.10032 · Cited by 583 · cited 3×
“For example, Kim et al (2006) use neural networks, decision tree and logistic regression to predict churn, using the result from the best performing model in the calculation of CLV. Neslin et al (2006) analyse the results from different data mining techniques used to calculate churn and conclude that logistic regressi…”
BackgroundDefection detection: measuring and understanding the predictive accuracy of customer churn models
Neslin, S. A., Gupta, S., Kamakura, W., Lu, J., & Mason, C. H. (2006) · Journal of Marketing Research · doi:10.1509/jmkr.43.2.204 · Cited by 574 · cited 3×
“…the result from the best performing model in the calculation of CLV. Neslin et al (2006) analyse the results from different data mining techniques used to calculate churn and conclude that logistic regression and decision trees performed well and are good techniques to be used when analysing churn.”
MethodCan we predict customer lifetime value?
Malthouse, E. C., & Blattberg, R. C. (2005) · Journal of Interactive Marketing · doi:10.1002/dir.20027 · Cited by 248
Larivière, B., & Van den Poel, D. (2004) · Expert Systems with Applications · doi:10.1016/j.eswa.2004.02.002 · Cited by 159
Cited by
Comprehensible classification models: a position paper
Freitas, A. A. (2014) · ACM SIGKDD Explorations Newsletter · doi:10.1145/2594473.2594475 · Cited by 572
Verbeke, W., Dejaeger, K., Martens, D., Hur, J., & Baesens, B. (2012) · European Journal of Operational Research · doi:10.1016/j.ejor.2011.09.031 · Cited by 462
“…regression (Lemmens and Croux, 2006; Neslin et al., 2006; Burez and Van den Poel, 2009), non-parametric statistical models like for instance k-nearest neighbor (Datta et al., 2000), decision trees (Wei and Chiu, 2002; Lima et al., 2009), and neural networks (Au et al., 2003; Hung et al., 2006).”
MethodSelf-citationBuilding comprehensible customer churn prediction models with advanced rule induction techniques
Verbeke, W., Martens, D., Mues, C., & Baesens, B. (2011) · Expert Systems with Applications · doi:10.1016/j.eswa.2010.08.023 · Cited by 363
Self-citationThe value of vehicle telematics data in insurance risk selection processes
Baecke, P., & Bocca, L. (2017) · Decision Support Systems · doi:10.1016/j.dss.2017.04.009 · Cited by 152
Why you should stop predicting customer churn and start using uplift models
Devriendt, F., Berrevoets, J., & Verbeke, W. (2021) · Information Sciences · doi:10.1016/j.ins.2019.12.075 · Cited by 90
Metadata
- Authors
- Lima, E., Mues, C., & Baesens, B.
- Venue
- Journal of the Operational Research Society
- Publisher
- Informa UK Limited
- Published
- 2009-08
- Volume
- 60
- Issue
- 8
- Pages
- 1096-1106
- ISSN
- 0160-5682 (ISSN-L)
- 1476-9360
- OpenAlex
- W1967465096
- Cited by
- 84
- References
- 6
- FWCI
- 13.52
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.