<![CDATA[Customer churn is a condition when customers stop using company services within a certain period. A high churn rate can negatively impact company revenue and customer loyalty, especially in telecommunications companies. This study aims to implement the Random Forest algorithm to predict customer churn in order to support customer retention strategies. The dataset used in this study is Telco Customer Churn obtained from Kaggle with a total of 7043 customer records. This research applies the CRISP-DM methodology consisting of business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The preprocessing stage includes handling missing values, transforming data types, and encoding categorical data. The modeling process uses the Random Forest Classifier algorithm with an 80:20 split between training and testing data. The results show that the Random Forest model achieved an accuracy of 77% in predicting customer churn. The most influential factors affecting churn based on feature importance are TotalCharges, tenure, MonthlyCharges, and customer contract types. Based on the research results, the Random Forest algorithm can help companies identify customers with churn potential and support more effective customer retention strategies.]]>
Copyrights © 2026
