Abstract :
Sentimen analysis of guest reviews plays a very important role in the hospitality business to understand customer responses to the services provided. This study aims to determine the process of guest review sentimen analysis using the Naive Bayes Classifier algorithm, the level of accuracy produced and a recommendation for hotels related to the results of the analysis. Guest review data was collected from the Traveloka website then went through a preprocessing process, namely text cleaning, casefolding, tokenization, stopword removal and stemming. The Naive Bayes Classifier algorithm is the preferred analysis model because of its simplicity and ability to handle text effectively. The results of the analysis show that the model used is able to provide an accuracy level of 91% which means it has a fairly good level of accuracy.