Examining the recommending behavior of guests with artificial intelligence in accommodation establishments

Customer information remains a key strategic point in hospitality management. However, the role that large amounts of existing knowledge can play in Guest Relationship Management (CRM) systems is still in its infancy when it is addressed using emerging machine learning techniques for efficient customer profiling. In this study, by using decision trees, the recommending behaviors of the guests were examined by using decision trees method by using the data in the guest relations department system of a branch of an international hotel chain operating in Belek. The recommending behavior was examined with two models. First, it was examined whether demographic variables affect the recommending behavior. It has been found that the loyalty program membership of the guests and the amount of spending for the holiday affect the recommendation behavior. The second model of the research was created to examine whether the evaluation of the services received by the guests during their stay affects the behavior of recommending the hotel. In this model, firstly, the recommendation behavior in general was examined and then the determinants of recommending behavior of the pandemic, gender, holiday type, nationality and booking through different channels were determined. Overall satisfaction was identified as a key determinant of recommending behavior in all models. Behavior determinants of each model are given in the relevant sections. The study is one of the rare studies examining the data in the CRM system. Previous studies generally analyzed the data collected by questionnaires with classical statistical methods. This study, on the other hand, obtained data from a hotel's CRM system and analyzed this data with decision trees from machine learning methods. In this respect, the study is an original work. The findings of the study give very important clues to the relevant hotel and to all practitioners in this field so that they can positively affect the behavior of guests recommending their hotel. It also provides a good guide for researchers in this field to use machine learning techniques in the related field.