Predicting Customer Churn in the Tourism Industry Using Machine Learning

Abstract

Customer churn, or service cancellation, poses a critical challenge that directly impacts the revenue and long-term sustainability of the tourism industry. This research aims to develop a proactive predictive system to assess customer cancellation risks by applying data analytics and Machine Learning techniques. Utilizing a travel behavior dataset comprising 954 records, the study evaluated and compared the performance of three predictive algorithms: Logistic Regression, Decision Tree, and Random Forest. The findings reveal that the Decision Tree model outperformed the others, achieving a robust prediction accuracy of 87%. This research provides a strategic advantage for tourism enterprises, enabling them to identify at-risk customers in advance, tailor proactive retention strategies, and effectively optimize their marketing resources.