Predicting Customer Churn in Travel and Tour Industry Using Machine Learning Algorithm Approaches

Abstract

Predicting customer churn in the airline sector of tour and travel poses unique challenges, necessitating advanced machine learning approaches to proactively tackle dissatisfaction, optimize service reliability, and fortify loyalty within fluctuating travel patterns and preferences. This study analyzed the application of machine learning algorithms to predict customer churn in the tour and travel industry. Leveraging data obtained from Kaggle, including factors like frequent flights, annual income, and social media engagement, the study employs various classifiers and attribute selection techniques to identify key predictors of churn. Through rigorous evaluation using five-fold cross-validation, the J48 decision tree classifier emerges as the most reliable model, achieving an accuracy of 84.53% and demonstrating good agreement. The findings underscore the potential of machine learning in enabling proactive customer retention strategies and enhancing business performance in the tour and travel sector.