A Smart Tourism Case Study: Classification of Accommodation Using Machine Learning Models Based on Accommodation Characteristics and Online Guest Reviews

Abstract

This paper deals with the analysis of data retrieved from a web page for booking accommodation. The main idea of the research is to analyze the relationship between accommodation factors and customer reviews in order to determine the factors that have the greatest influence on customer reviews. Machine learning methods are applied to the collected data and models that can predict the review category for those accommodations that are not evaluated by users are trained. The relationship between certain accommodation factors and classification accuracy of the models is examined in order to get detailed insight into the data used for model training, as well as to make the models more interpretable. The classification accuracy of each model is tested and the precision and recall of the models are examined and compared.