Who will be your next customer: A machine learning approach to customer return visits in airline services

Returning customer rates are a primary determinant of the success of a service; thus, both the motivations for and hindrances to return visits from customers are extensively investigated. Accordingly, this study aims to estimate the probability of customers’ return visits to airline services using a machine learning approach on the received feedback comments and satisfaction ratings regarding the previous usage of the service. By considering the sentimental features in the comments with seven classifiers, the results show an accuracy of 83.42% for predicting the customers’ return visits. Moreover, a higher word count of feedback written by the customers can lead to a higher degree of prediction accuracy. Based on these results, both the implications and limitations to customer preferences are presented.