Feature analysis and evaluation of hotel customer satisfaction based on decision tree

Abstract

The global hotel industry is in the process of rebounding from the impacts of the COVID-19 pandemic. Despite a surge in demand, traveler and customer satisfaction are experiencing a decline. To ensure the prosperity of hotels during this recovery phase, it becomes imperative to pinpoint the pivotal factors that dictate customer satisfaction. This research employs a decision tree algorithm for in-depth data analysis, mathematical modeling, and feature selection, all aimed at unveiling the determinants of passenger contentment. The model is subsequently refined and streamlined through the careful selection of the most impactful features. This study successfully identifies the paramount components that exert influence on customer satisfaction, with notable emphasis on factors such as hotel location, pricing, and booking procedures. Possessing insight into these determinants empowers hotels to concentrate their efforts on specific areas of enhancement, thus elevating the overall level of customer satisfaction. The implications of these findings hold practical significance for hotels, offering invaluable insights for the enhancement of service quality and customer contentment. Addressing the highlighted factors allows hotels to optimize operational efficiency, fortify their competitive edge, and secure enduring success in the post-pandemic recovery era. Moreover, by grasping customer preferences and augmenting satisfaction, hotels can safeguard their brand reputation and foster unwavering customer loyalty.