Machine learning based customer churn prediction in home appliance rental business

Abstract

Customer churn is a major issue for large enterprises. In particular, in the rental business sector, companies are looking for ways to retain their customers because they are their main source of revenue. The main contribution of our work is to analyze the customer behavior information of actual water purifier rental company, where customer churn occurs very frequently, and to develop and verify the churn prediction model. A machine learning algorithm was applied to a large-capacity operating dataset of rental care service in an electronics company in Korea, to learn meaningful features. To measure the performance of the model, the F-measure and area under curve (AUC) were adopted whereby an F1 value of 93% and an AUC of 88% were achieved. The dataset containing approximately 84,000 customers was used for training and testing. Another contribution was to evaluate the inference performance of the predictive model using the contract status of about 250,000 customer data currently in operation, confirming a hit rate of about 80%. Finally, this study identified and calculated the influence of key variables on individual customer churn to enable a business person (rental care customer management staff) to carry out customer-tailored marketing to address the cause of the churn.

Unleashing the Potentials of Artificial Intelligence for Micro, Small, and Medium Enterprises: A Systematic Literature Review

“…A. Customer Churn Prediction This sub-category has been extensively discussed in previous studies, with four studies addressing these topics [17], [18], [22], [25]. The urgency of customer churn remains relevant and continues to trend until this year, as evidenced by research published in 2023.…”

“…The studies cover supervised, unsupervised, and ensemble learning. In terms of quantity, research on ensemble learning is the most dominant, including algorithms such as Random Forest, Light Gradient Boosting Machine, Fuzzy Rule-based Classifier, Extreme Gradient Boosting, and Gradient Boosting [17], [18], [22], [27]. The prevalence of ensemble usage indicates that the research aims to maximize performance through the combination of various models.…”

“…customer churn using ensemble methods, Random Forest (RF), and Light Gradient-Boosting Machine (LightGBM)[17] Sales Q1(Faritha Banu et al, 2022) Model for predicting and managing customer churn and non-churn in the business sector using advanced selection, optimalization, and classification techniques deep learning-based method to enhance the classification accuracy of sentiment analysis on product comments[19] Sales Q1(Qiu & Chen, 2022) Model for analysing environmental costs using Back Propagation Neural Network (BPNN) optimized by the NSGA-II algorithm[20] Financial Q2(Chen & Long, 2023) Model for predicting financial risks based on a neural network with a combination of Factor Analysis (FA), Particle Swarm Optimization (PSO), and Long Short-Term Memory (LSTM) techniques[21] customer churn using Extreme Gradient Boosting (XGBoost) and Logistic Regression (LR)[22] analysis and goal extraction from customer feedback using a combination of Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN)[23] Sales Q2Li et al, 2022) Implementation of Probabilistic Neural Network (PNN) for environmental cost control classification and Particle Swarm Optimization (PSO) algorithm as the decision-making system[24] Financial Q3(J et al, 2023) Customer churn prediction using a Support Vector Machine algorithm and Hybrid Recommendation Strategy to prevent churn[25] financial risks using an optimized structure combining Temporal Convolutional Network-Long (TCN) with Long Short-Term Memory (LSTM)[26] Financial Q4Wisesa et al, 2020) Model for predicting sales in the B2B context using various machine learning algorithms, with Gradient Boost performing the best[27] Sales ~(Sánchez-Torres et al, 2022) …”

Using Machine Learning Algorithms to Analyze Customer Churn with Commissions Rate for Stocks in Brokerage Firms and Banks

“…For example, in one study in the literature [6] provides offer readers a general understanding of the commonly used data mining methods, their results, and performance, while also highlighting the need for further research in the Telecommunication Industry. In contrast, another study [14] the analysis involved examining the customer behavior data of a real water purifier rental service within an electronics company in Korea to create and validate a churn prediction model. The model's performance was assessed using the Fmeasure and area under curve (AUC) metrics.…”

An Order-Theoretic Study on Formal Concept Analysis

“…These strategies, ranging from enticing discounts and captivating loyalty rewards to intimately personalized incentives, are unified by the shared objective of staunchly preventing churn. By adopting this strategic and data-driven methodology, businesses not only gain in-into potential churn scenarios but also equip themselves to deploy effective and preemptive measures to enhance customer loyalty and overall satisfaction [16,17].…”