Measuring Hotel Service Quality in Borobudur Temple Using Opinion Mining

The increasing of visitors number to Borobudur Temple due to infrastructure development has pushed hospitality industry to meet the level of quality expected from various customer, especially for hotel services. Currently, customers are benefited to plan their stay by looking through the hotel reviews, yet this review could also be profitable for hotels. This paper attempted to gain insight from online reviews of the hospitality industry in the Borobudur Temple area by employing opinion mining. The objective of this study is to measure the sentiment of hotel services quality in Borobudur Temple area using hotel reviews based on HOLSERV Plus dimensions (Room, Facility, Surrounding, Employee, and Reliability), which are considered to be relevant to measure the quality of hotel services. The hotel reviews is classified into five dimensions of HOLSERV Plus based on three different algorithms; Naïve Bayes, Support Vector Machine, and k-Nearest Neighbor. The result showed that Support Vector Machine had the highest accuracy, precision, and F1 score compared to other algorithms. Moreover, all dimensions have positive sentiment on average where employee dimension has the highest positive sentiment compared to other dimensions.