4.7 Article

Ensemble Methods in Customer Churn Prediction: A Comparative Analysis of the State-of-the-Art

Related references

Note: Only part of the references are listed.
Article Computer Science, Artificial Intelligence

CATCHM: A novel network-based credit card fraud detection method using node representation learning*

Rafael Van Belle et al.

Summary: This paper proposes a network-based credit card fraud detection method called CATCHM, which leverages representation learning to improve fraud detection performance by avoiding manual feature engineering and explicitly considering the relational structure of transactions.

DECISION SUPPORT SYSTEMS (2023)

Article Operations Research & Management Science

B2Boost: instance-dependent profit-driven modelling of B2B churn

Bram Janssens et al.

Summary: This paper aims to enhance the current practices in business-to-business (B2B) customer churn prediction modelling by introducing a novel expected maximum profit measure and a gradient boosting classifier called B2Boost. By considering customer value and company profit, this study improves the current practices and shows significant expected maximal profit gains.

ANNALS OF OPERATIONS RESEARCH (2022)

Article Computer Science, Artificial Intelligence

Churn prediction in digital game-based learning using data mining techniques: Logistic regression, decision tree, and random forest

Mai Kiguchi et al.

Summary: Educational Technology (EdTech) is an industry that combines education and technological advancements. Digital game-based learning (DGBL) is a specific category within EdTech. This study proposes an approach for defining and predicting churn in DGBL by analyzing a dataset from a Japanese company. The results indicate the effectiveness of the approach in determining and predicting churn in DGBL.

APPLIED SOFT COMPUTING (2022)

Article Engineering, Electrical & Electronic

An Ensemble Framework for Improving the Prediction of Deleterious Synonymous Mutation

Na Cheng et al.

Summary: In this study, a new method called EnDSM was proposed to accurately predict deleterious synonymous mutations by combining multiple features and machine learning algorithms. The results showed that EnDSM outperformed other predictors on both training and testing datasets. This research is of great importance for the prediction of synonymous mutations in the field of medical genomics.

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY (2022)

Article Chemistry, Multidisciplinary

Customer Churn Prediction in B2B Non-Contractual Business Settings Using Invoice Data

Milan Mirkovic et al.

Summary: This paper presents an approach to predict customer churn in non-contractual B2B settings using invoice-level data. The results show that robust models for different churn definitions can be created using only invoice-level data. Using more historical data for feature engineering tends to improve the performance of some classifiers. It is also confirmed that the multi-slicing approach to dataset creation yields better performing models compared to the traditionally used single-slicing approach.

APPLIED SCIENCES-BASEL (2022)

Article Computer Science, Information Systems

A New 360° Framework to Predict Customer Lifetime Value for Multi-Category E-Commerce Companies Using a Multi-Output Deep Neural Network and Explainable Artificial Intelligence

Gulsah Yilmaz Benk et al.

Summary: Online purchasing has seen rapid growth due to its efficiency, convenience, low cost, and product variety. This has led to an increase in the number of online multi-category e-commerce retailers, requiring each company to optimize their business strategy. In addition to customer lifetime value (CLV), this study introduces distinct product category (DPC) and trend in amount spent (TAS) as novel factors to determine the most valuable customers. A new framework is proposed using a multi-output deep neural network (DNN) model to forecast CLV, DPC, and TAS. Customer clusters are constructed based on these variables, allowing for better segmentation of end-users. The results show that the multi-output DNN model outperforms other models and is more suitable for multi-category e-commerce retailers.

INFORMATION (2022)

Article Computer Science, Artificial Intelligence

Ensemble learning with a genetic algorithm for surface roughness prediction in multi-jet polishing

Ruoxin Wang et al.

Summary: This study developed a robust surface roughness prediction model based on ensemble learning with a genetic algorithm (ELGA) for predicting surface roughness in multi-jet polishing (MJP) of 3D-printed stainless steel. Experimental results showed that the ELGA model outperformed six other machine learning-based models.

EXPERT SYSTEMS WITH APPLICATIONS (2022)

Article Computer Science, Information Systems

A Sampling-Based Stack Framework for Imbalanced Learning in Churn Prediction

Soumi De et al.

Summary: Churn prediction is a popular research topic, with imbalanced class distribution often seen in related datasets. Researchers commonly use data-level solutions like over-sampling and under-sampling, and are exploring the integration of these solutions with advanced frameworks like ensembles to improve predictive performance.

IEEE ACCESS (2022)

Article Computer Science, Information Systems

Targeting customers for profit: An ensemble learning framework to support marketing decision-making

Stefan Lessmann et al.

Summary: The study highlights the importance of targeting the right customer groups in marketing campaigns and proposes a profit-conscious ensemble selection modeling framework that integrates statistical learning principles with business objectives, contributing to the field of profit analytics.

INFORMATION SCIENCES (2021)

Article Computer Science, Artificial Intelligence

Leveraging unstructured call log data for customer churn prediction

Nhi N. Y. Vo et al.

Summary: Customer retention is crucial in the financial services industry, and machine learning has been used to predict client churn risks. While existing approaches mainly rely on structured data, mining unstructured data can provide more insights. The research introduced a model utilizing spoken contents in phone communication for customer churn prediction, which showed promising results in accurately predicting risks and generating meaningful insights.

KNOWLEDGE-BASED SYSTEMS (2021)

Article Computer Science, Artificial Intelligence

Predicting donation behavior: Acquisition modeling in the nonprofit sector using Facebook data

Lisa Schetgen et al.

Summary: This study demonstrates the value of Facebook data in predicting first-time donation behavior and acquiring new donors for nonprofit organizations. The combination of singular value decomposition and logistic regression outperformed other analytical methodologies, with Facebook pages and categories being the most important data types. Factors related to age, education, residence, and other dimensions played a significant role in predicting donation behavior.

DECISION SUPPORT SYSTEMS (2021)

Article Computer Science, Artificial Intelligence

Box office sales and social media: A cross-platform comparison of predictive ability and mechanisms

Matthias Bogaert et al.

Summary: The research shows that social media data significantly increases the predictive power of traditional box office prediction models. Facebook data outperforms Twitter data and including user-generated content consistently improves predictive power. Combination variables based on volume and valence of Facebook comments are identified as the most important variables.

DECISION SUPPORT SYSTEMS (2021)

Article Computer Science, Artificial Intelligence

Spline-rule ensemble classifiers with structured sparsity regularization for interpretable customer churn modeling

Koen W. De Bock et al.

Summary: Customer retention management relies heavily on advanced statistical and machine learning algorithms, with customer churn prediction being a crucial tool. This study introduces rule ensembles and spline-rule ensembles as promising classification algorithms for customer churn prediction, aiming to improve predictive performance and interpretability through structured regularization.

DECISION SUPPORT SYSTEMS (2021)

Article Management

Deep learning for credit scoring: Do or don't?

Bjorn Rafn Gunnarsson et al.

Summary: Developing accurate credit scoring models is crucial for financial institutions, with various classification algorithms proposed. However, the use of deep learning algorithms in credit scoring has been overlooked. This research compared the performance of deep neural networks with traditional and ensemble methods, finding that XGBoost outperformed all other methods, indicating that deep learning algorithms may not be suitable for credit scoring.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2021)

Proceedings Paper Computer Science, Artificial Intelligence

Churn Prediction in Telecom Industry using Machine Learning Ensembles with Class Balancing

Abdullahi Chowdhury et al.

Summary: Telecommunication service providers are facing intense competition and the need to accurately analyze and predict customer churn behavior. Data imbalance presents a major challenge in building better prediction models, with the model using sophisticated oversampling techniques and ensemble methods showing improved performance.

2021 IEEE ASIA-PACIFIC CONFERENCE ON COMPUTER SCIENCE AND DATA ENGINEERING (CSDE) (2021)

Proceedings Paper Computer Science, Information Systems

Leveraging the efficiency of Ensembles for Customer Retention

Neha Sunil Bhujbal et al.

Summary: Banks need to stay updated on market offers and customer loyalty to survive in the competitive industry. Key parameters for analyzing customer behavior include demographics and credit card usage details. Models for predicting customer churn often use machine learning algorithms or ensemble algorithms.

PROCEEDINGS OF THE 2021 FIFTH INTERNATIONAL CONFERENCE ON I-SMAC (IOT IN SOCIAL, MOBILE, ANALYTICS AND CLOUD) (I-SMAC 2021) (2021)

Article Management

How training on multiple time slices improves performance in churn prediction

Theresa Gattermann-Itschert et al.

Summary: Customer churn prediction models using multi-slicing training approach, which trains models on multiple time slices of data, have shown significant improvement in prediction performance compared to traditional single-slice training. By addressing the rarity of churn samples and potential overfitting, multi-slicing makes the model more generalizable, especially in scenarios where conditions fluctuate over time.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2021)

Article Computer Science, Information Systems

Integrated Churn Prediction and Customer Segmentation Framework for Telco Business

Shuli Wu et al.

Summary: In the telco industry, churn management has become crucial, and the integration of churn prediction and customer segmentation is relatively scarce. This paper proposes an integrated customer analytics framework that aims to better manage customer churn by processing, analyzing, and predicting data, enabling more precise retention strategies to be adopted.

IEEE ACCESS (2021)

Article Biochemical Research Methods

Prediction of Neuropeptides from Sequence Information Using Ensemble Classifier and Hybrid Features

Yannan Bin et al.

JOURNAL OF PROTEOME RESEARCH (2020)

Article Computer Science, Information Systems

A Survey on Churn Analysis in Various Business Domains

Jaehyun Ahn et al.

IEEE ACCESS (2020)

Proceedings Paper Computer Science, Artificial Intelligence

Chum Prediction in Telecommunication using Logistic Regression and Logit Boost

Hemlata Jain et al.

INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND DATA SCIENCE (2020)

Article Computer Science, Artificial Intelligence

A novel tree-based dynamic heterogeneous ensemble method for credit scoring

Yufei Xia et al.

EXPERT SYSTEMS WITH APPLICATIONS (2020)

Article Computer Science, Interdisciplinary Applications

Fitting Prediction Rule Ensembles with R Package pre

Marjolein Fokkema

JOURNAL OF STATISTICAL SOFTWARE (2020)

Article Management

Evaluating multi-label classifiers and recommender systems in the financial service sector

Matthias Bogaert et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2019)

Article Business

Customer churn prediction in telecommunication industry using data certainty

Adnan Amin et al.

JOURNAL OF BUSINESS RESEARCH (2019)

Article Operations Research & Management Science

Evaluating the importance of different communication types in romantic tie prediction on social media

Matthias Bogaert et al.

ANNALS OF OPERATIONS RESEARCH (2018)

Article Management

Benchmarking sampling techniques for imbalance learning in churn prediction

Bing Zhu et al.

JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY (2018)

Article Management

A new hybrid classification algorithm for customer churn prediction based on logistic regression and decision trees

Arno De Caigny et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2018)

Article Computer Science, Artificial Intelligence

A novel heterogeneous ensemble credit scoring model based on bstacking approach

Yufei Xia et al.

EXPERT SYSTEMS WITH APPLICATIONS (2018)

Article Computer Science, Artificial Intelligence

Social network analytics for churn prediction in telco: Model building, evaluation and network architecture

Maria Oskarsdottir et al.

EXPERT SYSTEMS WITH APPLICATIONS (2017)

Article Computer Science, Artificial Intelligence

A comparative study on base classifiers in ensemble methods for credit scoring

Joaquin Abelian et al.

EXPERT SYSTEMS WITH APPLICATIONS (2017)

Article Computer Science, Information Systems

An empirical comparison of techniques for the class imbalance problem in churn prediction

Bing Zhu et al.

INFORMATION SCIENCES (2017)

Article Computer Science, Artificial Intelligence

A comparative analysis of data preparation algorithms for customer churn prediction: A case study in the telecommunication industry

Kristof Coussement et al.

DECISION SUPPORT SYSTEMS (2017)

Article Management

Social media optimization: Identifying an optimal strategy for increasing network size on Facebook

Michel Ballings et al.

OMEGA-INTERNATIONAL JOURNAL OF MANAGEMENT SCIENCE (2016)

Article Computer Science, Artificial Intelligence

Including high-cardinality attributes in predictive models: A case study in churn prediction in the energy sector

Julie Moeyersoms et al.

DECISION SUPPORT SYSTEMS (2015)

Article Management

CRM in social media: Predicting increases in Facebook usage frequency

Michel Ballings et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2015)

Article Management

Benchmarking state-of-the-art classification algorithms for credit scoring: An update of research

Stefan Lessmann et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2015)

Article Computer Science, Artificial Intelligence

A Bayesian approach for comparing cross-validated algorithms on multiple data sets

Giorgio Corani et al.

MACHINE LEARNING (2015)

Article Computer Science, Interdisciplinary Applications

A comparison of machine learning techniques for customer churn prediction

T. Vafeiadis et al.

SIMULATION MODELLING PRACTICE AND THEORY (2015)

Article Computer Science, Artificial Intelligence

Particle swarm optimisation for feature selection in classification: Novel initialisation and updating mechanisms

Bing Xue et al.

APPLIED SOFT COMPUTING (2014)

Article Computer Science, Artificial Intelligence

A survey of multiple classifier systems as hybrid systems

Michal Wozniak et al.

INFORMATION FUSION (2014)

Article Computer Science, Artificial Intelligence

Integrating complementary techniques for promoting diversity in classifier ensembles: A systematic study

Diego S. C. Nascirnento et al.

NEUROCOMPUTING (2014)

Article Computer Science, Artificial Intelligence

Kernel Factory: An ensemble of kernel machines

Michel Ballings et al.

EXPERT SYSTEMS WITH APPLICATIONS (2013)

Article Computer Science, Artificial Intelligence

A Novel Profit Maximizing Metric for Measuring Classification Performance of Customer Churn Prediction Models

Thomas Verbraken et al.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2013)

Article Computer Science, Artificial Intelligence

Ensemble classifier generation using non-uniform layered clustering and Genetic Algorithm

Ashfaqur Rahman et al.

KNOWLEDGE-BASED SYSTEMS (2013)

Article Business

Customer churn prediction in the online gambling industry: The beneficial effect of ensemble learning

Kristof Coussement et al.

JOURNAL OF BUSINESS RESEARCH (2013)

Article Computer Science, Hardware & Architecture

Churn prediction in telecom using Random Forest and PSO based data balancing in combination with various feature selection strategies

Adnan Idris et al.

COMPUTERS & ELECTRICAL ENGINEERING (2012)

Article Management

New insights into churn prediction in the telecommunication sector: A profit driven data mining approach

Wouter Verbeke et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2012)

Article Computer Science, Artificial Intelligence

Uniformly subsampled ensemble (USE) for churn management: Theory and implementation

Namhyoung Kim et al.

EXPERT SYSTEMS WITH APPLICATIONS (2012)

Article Computer Science, Artificial Intelligence

Customer event history for churn prediction: How long is long enough?

Michel Ballings et al.

EXPERT SYSTEMS WITH APPLICATIONS (2012)

Article Computer Science, Artificial Intelligence

Reconciling performance and interpretability in customer churn prediction using ensemble learning based on generalized additive models

Koen W. De Bock et al.

EXPERT SYSTEMS WITH APPLICATIONS (2012)

Article Management

A probability-mapping algorithm for calibrating the posterior probabilities: A direct marketing application

Kristof Coussement et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2011)

Article Computer Science, Artificial Intelligence

Credit card churn forecasting by logistic regression and decision tree

Guangli Nie et al.

EXPERT SYSTEMS WITH APPLICATIONS (2011)

Article Computer Science, Artificial Intelligence

An empirical evaluation of rotation-based ensemble classifiers for customer churn prediction

Koen W. De Bock et al.

EXPERT SYSTEMS WITH APPLICATIONS (2011)

Article Computer Science, Artificial Intelligence

Memetic Algorithms for Continuous Optimisation Based on Local Search Chains

Daniel Molina et al.

EVOLUTIONARY COMPUTATION (2010)

Article Computer Science, Artificial Intelligence

Handling class imbalance in customer churn prediction

J. Burez et al.

EXPERT SYSTEMS WITH APPLICATIONS (2009)

Article Computer Science, Artificial Intelligence

Customer churn prediction by hybrid neural networks

Chih-Fong Tsai et al.

EXPERT SYSTEMS WITH APPLICATIONS (2009)

Article Computer Science, Artificial Intelligence

Customer churn prediction using improved balanced random forests

Yaya Xie et al.

EXPERT SYSTEMS WITH APPLICATIONS (2009)

Article Computer Science, Artificial Intelligence

Churn prediction in subscription services: An application of support vector machines while comparing two parameter-selection techniques

Kristof Coussement et al.

EXPERT SYSTEMS WITH APPLICATIONS (2008)

Article Computer Science, Artificial Intelligence

CRM at a pay-TV company: Using analytical models to reduce customer attrition by targeted marketing for subscription services

Jonathan Burez et al.

EXPERT SYSTEMS WITH APPLICATIONS (2007)

Article Computer Science, Artificial Intelligence

Applying data mining to telecom chum management

SY Hung et al.

EXPERT SYSTEMS WITH APPLICATIONS (2006)

Article Management

Using simulated annealing to optimize the feature selection problem in marketing applications

R Meiri et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2006)

Article Business

Bagging and boosting classification trees to predict churn

A Lemmens et al.

JOURNAL OF MARKETING RESEARCH (2006)

Article Computer Science, Artificial Intelligence

Logistic model trees

N Landwehr et al.

MACHINE LEARNING (2005)

Article Management

Customer attrition analysis for financial services using proportional hazard models

D Van den Poel et al.

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH (2004)

Article Management

Benchmarking state-of-the-art classification algorithms for credit scoring

B Baesens et al.

JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY (2003)

Article Computer Science, Artificial Intelligence

Attribute bagging: improving accuracy of classifier ensembles by using random feature subsets

R Bryll et al.

PATTERN RECOGNITION (2003)

Article Computer Science, Artificial Intelligence

Ensembling neural networks: Many could be better than all

ZH Zhou et al.

ARTIFICIAL INTELLIGENCE (2002)

Article Computer Science, Interdisciplinary Applications

Stochastic gradient boosting

JH Friedman

COMPUTATIONAL STATISTICS & DATA ANALYSIS (2002)

Article Computer Science, Artificial Intelligence

Random forests

L Breiman

MACHINE LEARNING (2001)