[Paper Review] Customer Churn Prediction Model using Explainable Machine Learning

Jitendra Maan, Harsh Maan|arXiv (Cornell University)|Mar 2, 2023

Customer churn and segmentation8 citations

TL;DR

The paper develops a customer churn prediction model using explainable ML, identifying XGBoost as the most effective classifier and proposing a Shapley-value based approach to explain feature importance.

ABSTRACT

It becomes a significant challenge to predict customer behavior and retain an existing customer with the rapid growth of digitization which opens up more opportunities for customers to choose from subscription-based products and services model. Since the cost of acquiring a new customer is five-times higher than retaining an existing customer, henceforth, there is a need to address the customer churn problem which is a major threat across the Industries. Considering direct impact on revenues, companies identify the factors that increases the customer churn rate. Here, key objective of the paper is to develop a unique Customer churn prediction model which can help to predict potential customers who are most likely to churn and such early warnings can help to take corrective measures to retain them. Here, we evaluated and analyzed the performance of various tree-based machine learning approaches and algorithms and identified the Extreme Gradient Boosting XGBOOST Classifier as the most optimal solution to Customer churn problem. To deal with such real-world problems, Paper emphasize the Model interpretability which is an important metric to help customers to understand how Churn Prediction Model is making predictions. In order to improve Model explainability and transparency, paper proposed a novel approach to calculate Shapley values for possible combination of features to explain which features are the most important/relevant features for a model to become highly interpretable, transparent and explainable to potential customers.

Motivation & Objective

• Address the challenge of predicting and retaining customers in a digitized subscription-based market.
• Evaluate tree-based machine learning models for churn prediction to identify the most effective approach.
• Emphasize model interpretability to explain predictions to stakeholders.
• Propose a novel Shapley-value based method to explain feature contributions and improve transparency.

Proposed method

• Compare tree-based machine learning approaches for churn prediction.
• Identify Extreme Gradient Boosting (XGBoost) as the optimal classifier among evaluated models.
• Develop a novel approach to compute Shapley values for combinations of features to explain model predictions.
• Focus on interpretability and transparency to help stakeholders understand churn predictions.

Experimental results

Research questions

• RQ1 Which machine learning model provides the best performance for churn prediction among tree-based methods?
• RQ2 How can we enhance the explainability and transparency of churn predictions for potential customers?
• RQ3 Which features are most influential in predicting churn according to a Shapley-value based explanation?
• RQ4 Can a Shapley-value based approach effectively explain combinations of features in the churn model?

Key findings

• XGBoost classifier identified as the most optimal solution for churn prediction.
• A novel Shapley-value based approach is proposed to explain the impact of possible feature combinations on predictions.
• The study emphasizes model interpretability to make churn predictions comprehensible to customers and stakeholders.