Opinion Mining Of Tripadvisor Review Towards Five-star Hotels In Bandung City

Abstrak

New ways how and where customers look for recommendation and reviews have emerged. People nowadays tend to check the review and compare potential hotels before they go to make sure whether the hotel is suitable with their preferences or not. Travellers’ reviews, in the form of User Generated Content (UGC), have created more effective assessment system in comparison to the more traditional classifications such as hotel’s review in guidebooks. TripAdvisor as a market leader of review platforms represents the largest collection of UGC. It transforms not only how customers and hotel interact, but also the use of a new approach to knowledge as a source for customer insight. However, it is difficult for web users to read and understand contents due to the large number of reviews. The hotel management also finds it difficult to obtain any review to measure their service quality. This research aims to provide a comprehensive view of service quality for five-star hotels in Bandung city based on UGC. The dataset of online review is obtained from TripAdvisor. The author uses opinion mining technique to decompose user reviews into five dimensions to measure hotel service quality. This technique can help the hotel management to turn mountains of text opinion into fresh customer insights. The result of this research aims to provide an overview of five-star hotels in Bandung city what dimension of service quality that require attention to be improved. This overview can reduce the time required in understanding review contents for both the hotel management and web users. The overview indicates that each hotel has different dimensions that need more attention. Crowne Plaza, GH Universal, Hilton, InterContinental, Padma, Sheraton, and Trans have the least positive result for assurance in comparison to other dimensions, while Aryaduta, Novotel, and Papandayan have the least positive result for tangibles. This can be used for each hotel management to get customer insights that will help them deliver greater value for customers. Keywords: Service Quality, Multi-class Classification, Opinion Mining, Review Hotel