Customer Churn Prediction

Customer Churn Prediction

Thorat Anushka Prashant, Patil Saish Subash, Sayyed Sufiyaan Imtiyaz,

Sonavane Dhruv Suhas

anushkathorat6@gmail.com (Diploma last year KKW),

saishp83@gmail.com (Diploma last year KKW),

sayyedsufiyaan875@gmail.com (Diploma last year KKW),

dhruvsonavane45@gmail.com (Diploma last year KKW)

Abstract - Customer churn is a significant concern in the telecom industry as retaining existing customers is more cost-effective than acquiring new ones. Predicting customer churn is crucial for telecom companies to improve retention strategies. This study focuses on developing a churn prediction model using machine learning techniques to help telecom operators identify customers likely to churn. The model utilizes a large dataset containing customer information such as demographics, services, and payment methods. The model was built and tested in a Spark environment. Three machine learning algorithms were implemented: Decision Tree, Random Forest, and XGBoost. The goal is to assist telecom operators in predicting and managing customer churn, ultimately improving customer retention and business performance.