Exploring Customer Perceptions through Sentiment Analysis of Google Reviews at Rainbow Alamanda: SVM vs Naive Bayes Algorithm

Abstrak

As one of the popular family tourist destinations, Rainbow Alamanda Park has received thousands of reviews from visitors on the Google Review platform. These reviews reflect public perceptions of the quality of services and facilities offered, making it important to analyze them systematically. This study aims to analyze the sentiment of visitor reviews on Google Review regarding Rainbow Alamanda using two machine learning algorithms: Naive Bayes and Support Vector Machine (SVM), and to compare the performance of both methods. The research process follows the SEMMA approach (Sample, Explore, Modify, Model, Assess), utilizing a dataset of 2,394 reviews collected through web scraping techniques. The evaluation results show that the Naive Bayes method performed best with a training-to-testing data ratio of 70:30, achieving an accuracy of 86.32%, precision of 86.83%, recall of 85.81%, and an F1-score of 86.08%. Meanwhile, the SVM method with an RBF kernel (C=10, γ=0.1) achieved higher performance, with an accuracy of 88.44%, precision of 90.27%, recall of 88.31%, and an F1-score of 89.28%.

Identifier

• DOI: https://doi.org/10.26740/jdbim.v5i1.73320
• OpenAlex: https://openalex.org/W7172217256