Classifying Hotel Reviews into Criteria for Review Summarization

Recently, we can refer to user reviews in the shopping or hotel reservation sites. However, with the exponential growth of information of the Internet, it is becoming increasingly difficult for a user to read and understand all the materials from a large-scale reviews. In this paper, we propose a method for classifying hotel reviews written in Japanese into criteria, e.g., location and facilities. Our system firstly extracts words which represent criteria from hotel reviews. The extracted words are classified into 12 criteria classes. Then, for each hotel, each sentence of the guest reviews is classified into criterion classes by using two different types of Naive Bayes classifiers. We performed experiments for estimating accuracy of classifying hotel review into 12 criteria. The results showed the effectiveness of our method and indicated that it can be used for review summarization by guest’s criteria.