THE UTILIZATION OF ONLINE REVIEWS FOR HOTEL CONSUMER KANO MODEL(CASE: THREE-STAR HOTELS IN BANDUNG)

• Main Author: Adriana Widyanti, Karin
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/70461
• Institution: Institut Teknologi Bandung

One of the most important key success factors for a business is undoubtedly their customer’s satisfaction. The customer satisfaction can be modelled by the Kano model to evaluate the business’ performance in the eyes of their customers. However, the conventional Kano modelling method receives inputs in the form of Kano questionnaires that would require extensive resources of time, money, and effort to gather and process the questionnaires, thus very challenging to be implemented. A more modern alternative of this is to use online reviews that are publicly accessible on the internet as the input for the Kano model. Hospitality or hotel industry is one of the business sectors that prominently utilizes and depends on online reviews to evaluate their business performances. However, hotels have challenges in processing online reviews due to its limited ability to manually process online reviews. This research aims to solve those challenges for the hospitality industry by utilizing online reviews as the input for the Kano model and to develop a method for the data to be automatically processed, ultimately enabling the business to be able to evaluate their feature performances based on the reviews. The proposed method processes the online reviews data through three main steps. The first step is feature extraction and sentiment analysis with aspect-based sentiment analysis model, then followed with the measurement of the consumer sentiment effect to the customer satisfaction using a neural network model. The third and last step is identifying the Kano categories from the data using effectbased Kano model (EKM). This research focuses on 3-starred hotels located in Bandung as the case research, where 6,037 hotel reviews were collected from a hotel rating website, pegipegi.com. The feature extraction and sentiment analysis resulted in 18 features with the customer sentiment for each feature. The extracted features are AC, bathroom, bathroom facility, bedding, bedroom, bedroom facility, cleanliness, comfort, location, hotel facility, swimming pool, food & restaurants, reception, security staff, general staff, price, parking lot, and WiFi. Furthermore, the weights for positive sentiments (WiPos) and negative sentiments (WiNeg) were calculated for each feature and used as the input for building the EKM. The endresult of this final project is the EKM of the features where it was inferred that one feature, bathroom facilities, was categorized as Kano must-be, 17 features were categorized as Kano performance, and no features were categorized as Kano excitement. By the completion of this research, it is hoped that the work can help enable the hospitality industry to evaluate their performance using online reviews and identify the priorities of features that need to be improved immediately, without the high manual work and exhausting their resources. This research provide suggestions for the management of hospitality industry that are interested in implementing this proposed method of online review data processing to their business.