Leveraging Machine Learning for Sentiment Analysis in Hotel Applications: A Comparative Study of Support Vector Machine and Random Forest Algorithms

Abstract

This research aims to conduct sentiment analysis on user reviews of hotel booking applications such as Trivago, Tiket, Booking, Traveloka, and Agoda, collected from the Google Play Store. The dataset used consists of 5,000 user reviews, with 80% of the data allocated for training and 20% for testing. Two algorithms applied in this study are Support Vector Machine (SVM) and Random Forest, with performance evaluation based on accuracy, precision, recall, and F1-score metrics. The test results show that the Random Forest algorithm delivers the best performance on the Trivago application with 94% accuracy, 94% precision, 100% recall, and a 97% F1-score. Random Forest proves to be more effective in handling diverse review data, while the Support Vector Machine (SVM) algorithm also produces good results in sentiment classification. This research contributes to the development of sentiment analysis based on user reviews, which can be utilized by app developers and hotel management to improve service quality and user experience.