Review Machine Learning Models for Managing Hotel Cancellations in the Tourism Industry

This survey paper is one-stop-shop for keeping up with the latest research on predicting hospitality booking cancellations using machine learning and deep learning techniques. With cancellations having such a significant impact on hospitality revenue and reputation, it’s crucial to develop effective prediction models, and this paper highlights the various techniques used in recent studies. But accurate and reliable cancellation forecasting is more than just a theoretical exercise; it’s essential for successful revenue management in the tourism and hospitality industry. The paper delves into the challenges associated with demand forecasting, particularly in the context of last-minute bookings and changes in customer behavior. To help navigate these challenges, this survey paper suggests the neural network method as a promising approach for predicting reservation cancellations, enabling hospitality management to make informed decisions about which reservations to accept or reject. Although cancellation policies have been put in place by hospitality management to mitigate the effects of cancellations, issues such as demand and reputation persist. The paper underscores the significance of segregating data sets and feature engineering in predictive models, which is supported by the studies reviewed. Moreover, the paper examines the importance of extracting insights from extensive databases and leveraging predictive analytics to identify the factors that influence the likelihood of booking cancellations.