Claim Missing Document
Check
Articles

Found 1 Documents
Search

Sentiment Analysis of Lombok Tourism Destinations with Automatic Labeling Using Bidirectional Encoder Representations from Transformers (BERT) and Naïve Bayes Nora Ananda Putri; Lalu Mutawalli; Maulana Ashari
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 2 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Oktober 2026 (In Press)
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i2.28494

Abstract

Reviews on Google Maps can be used to determine travelers' perceptions of the destination, but the large amount of data makes manual analysis less effective. This study aims to analyze the sentiment of tourist reviews of beach destinations on the island of Lombok using BERT and Naïve Bayes. The data was obtained through Google Maps scraping and generated 12,040 data after the cleaning process. The research stages include preprocessing, translation into English, sentiment labeling using BERT, TF-IDF feature extraction and data splitting, and classification using Naïve Bayes. The results showed that positive sentiment dominated with 8569 data (72.03%), followed by neutral sentiment as many as 1,824 data (15.33%) and negative sentiment as many as 1,504 data (12.64%). The Naïve Bayes model obtained an accuracy of 76.13% and showed a fairly good performance, although it was more optimal in classifying positive sentiment than neutral and negative sentiment. Overall, the results of the study show that beach destinations on Lombok Island have a positive image in the eyes of tourists and can be an evaluation material for managers in improving the quality of tourism services and facilities.