การวิเคราะห์และแสดงมโนทัศน์ของพฤติกรรมของนักท่องเที่ยวที่มีต่อบริการโรงแรมในจังหวัดเชียงใหม่ โดยใช้เทคนิคการวิเคราะห์ความรู้สึก

Abstract

This independent study is learning to understand the feelings and behaviors of tourists affecting hostel accommodation services in a case study of Muang district, Chiang Mai province. The study focuses on a case study of the Muang district in Chiang Mai province. The data used in this research was collected from TripAdvisor.com, where 5,108 messages were crawled and separated into 17,092 sentences. The researcher will select a sample of entrepreneurs who operate hostel accommodations and conduct in-depth interviews with them. The aim is to gain insights into the various aspects of service that these entrepreneurs take into account when managing their businesses. The sample selection will be based on the characteristics of the sample group, in accordance with the research objectives. The sample group will consist of hostel accommodation entrepreneurs located in the Mueang Chiang Mai district of Chiang Mai province, who have received a certificate of excellence from TripAdvisor.com - a booking agent website and a reputable source for a wide variety of accommodations. The certificate of excellence from TripAdvisor.com is awarded to hospitality businesses that consistently provide exceptional service and is a recognition of excellence selected from around the world. Only approximately 10% of all businesses on TripAdvisor have received this award. The researcher employed this criterion to carefully select a sample group of hostel entrepreneurs who possess a comprehensive understanding of the hospitality industry. This was done to ensure that the sample obtained is qualified and can provide valuable insights into the hostel business, and to identify specific aspects of service that are important for hostels located in the Mueang Chiang Mai district of Chiang Mai province. These aspects will be used as a classification criterion for developing a sentiment analysis system that analyzes reviews of hostels written by travelers or guests. The Aspects-based Sentiment Analysis tool, which uses both Support Vector Machine (SVM) algorithm and the Multinomial Naïve Bayes algorithm (MultinomialNB), will be utilized in this study, and the accuracy scores will be compared to determine which algorithm performs better. Additionally, the study will investigate the business development of hostels by conducting a text-mining analysis of the reviews. According to the study, a system developed using the Support Vector Machine (SVM) algorithm can classify aspects of the hostel's services with the accuracy score is 93.27 percent, which is gave the highest accuracy score comparing with the Multinomial Naïve Bayes algorithm (MultinomialNB) with the accuracy score is 81.75 percent. In conclusion, based on the results of the study, it can be inferred that the support vector machine (SVM) model is a suitable approach for classifying customer review messages into various aspects of hostel service. The model demonstrated a satisfactory level of performance and accuracy, which provides confidence in its effectiveness for this purpose.