Harmonizing Sentiments: User Reviews from Social Media of Kuta Beach as a Tourist Attraction

Google Maps as a type Social Media is a popular navigation application created by Google. Since the market was very big and there are around 17.000 islands spread across the country, Indonesians use this application alot. Almost every tourist destination and place is here. Bali was one of the most visited places of Indonesia, reaching an outstanding 6,311,292 people and 36.91% of them were tourists. At Bali, Kuta Beach was one of the most visited and one of the most popular places in Bali. At the time of writing this research, the total number of reviews at Kuta Beach from Google Maps reached 39.638 reviews in total and rated 4.5 stars overall. This sentiment analysis research will be conducted by applying various classification models, such as Naive Bayes, Support Vector Machine, and Random Forest. The research will begin with a data collection process, where web scraping will be implemented to extract the sample data. The data will then undergo a Data and Text process, involving labeling, case folding, and various data cleaning methods, including normalization, removal of stopwords, tokenization, and stemming. The processed sample data will be divided in an 80:20 ratio, with 80% of the data used for training and 20% for testing. Following this, the data will be applied to the classification models. The results of the various classification models will be compared to identify which one among the three has the highest accuracy. Through this research, it was found that Random Forest has the highest accuracy rate, reaching 89%.