Building comprehensible customer churn prediction models with advanced rule induction techniques

By Wouter Verbeke, David Martens, C. Mues, Bart Baesens. 2011

372 citations recorded by the source corpus.

Subfields: Information Systems, Marketing, Organizational Behavior and Human Resource Management

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Marketing

In Organizational Behavior and Human Resource Management

Citations in the corpus

A Survey of Methods for Explaining Black Box Models (2018, background): “The degree of trust on a model increases if the model is built by respecting constraints of monotonicity given by the users [77, 87, 123].”
Falling Rule Lists (2014, background): “…context dependent (see Kodratoﬀ, 1994; Pazzani, 2000; Freitas, 2014; Huys-mans et al. , 2011; Allahyari & Lavesson, 2011; Martens & Baesens, 2010; R¨uping, 2006; Verbeke et al. , 2011; Martens et al. , 2011), and no matter how one measures it in one domain, it can be diﬀerent in the next domain.”
Learning Classification Models of Cognitive Conditions from Subtle Behaviors in the Digital Clock Drawing Test (2015, background): “…et al. 2007), crime prediction (Steinhart 2006; Andrade 2009; Ridgeway 2013), national defense (ABS Consulting 2002), marketing (Hauser et al. 2010; Verbeke et al. 2011; Wang et al. 2015), medical diagnosis (Tian and Tibshirani 2011; Van Belle et al. 2012; Letham et al. 2015; Wang and Rudin 2015),…”
Methods and Models for Interpretable Linear Classification (2014, background): “In turn, recent approaches to interpretable predictive modeling have sought to produce models with monotonicity constraints so that the relationship between input variables and the predicted response value matches the views of domain experts [10, 87, 114, 77].”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “Another method based on an ACO algorithm, called AntMinter+ [270], constructs monotonic rulesets by allowing the inclusion of domain knowledge via the definition of a directed acyclic graph representing the solution space.”
Churn prediction in new users of Yahoo! answers (2012, background): “Specifically, we would like to identify users that are about to quit, a task known as churn prediction [12, 15, 5, 10, 13, 18, 9].”
Characterization of Symbolic Rules Embedded in Deep DIMLP Networks: A Challenge to Transparency of Deep Learning (2017, methodology): “Cherkassky and Dhar generated rules from SVMs with the ALBA algorithm [34] for digits 5 and 8.”
Comprehensible classification models: a position paper (2014, background): “Overall, however, it is usually agreed that representations like decision trees, classification rules or decision tables tend to be more comprehensible to most users than mathematical equations (like linear combinations of attributes in oblique decision trees) and non-linear models like ANN or SVM [”
APATE: A novel approach for automated credit card transaction fraud detection using network-based extensions (2015, background): “Contents lists available at ScienceDirect Decision Support Systems j ourna l homepage: www.e lsev ie r .com/ locate /dss APATE: A novel approach for automated credit card transaction fraud detection using network-based extensions Véronique Van Vlasselaer a, Cristián Bravo b,⁎, Olivier Caelen c, Tina”
New insights into churn prediction in the telecommunication sector: A profit driven data mining approach (2012, methodology): “For an extensive literature review on customer churn prediction modeling one may refer to Verbeke et al. (2011).”
A Survey on the Explainability of Supervised Machine Learning (2020, background): “Furthermore, it is of great interest for banks and insurance companies to predict and understand customer churn to be able to develop a reasoned counteracting plan (due to high costs of seeking new customers) (Verbeke et al., 2011).”
A Churn Prediction Model Using Random Forest: Analysis of Machine Learning Techniques for Churn Prediction and Factor Identification in Telecom Sector (2019, background): “This is due to the cost involved in advertisements, workforce, and concessions which can scale up to almost ﬁve to six times than retaining existing customers [3].”
A Bayesian Approach to Learning Scoring Systems (2015, background): “, 2007; Martens, Baesens, and Van Gestel, 2009), churn analysis (Lima, Mues, and Baesens, 2009; Verbeke et al., 2011) to crime prediction (Ridgeway, 2013; Andrade, 2009), often a model can only be defensible if it is interpretable, and easy to use without a calculator (see Giraud-Carrier, 1998; Frei”