<![CDATA[This study focuses on conducting aspect-based sentiment analysis and topic modelling of tourism destinations in Indonesia, which are Borobudur Temple and Prambanan Temple using Google Maps and Tripadvisor user reviews. Aspect-based sentiment analysis was done using five classical machine learning algorithms, which are NaïveBayes (NB), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), and Extra Trees (ET) using the unigram+bigram+trigram feature and the application of combination of the use of training and test data, stopwords removal, stemming, emoji processing, and over-sampled training data. The performance of models was evaluated by comparing F1-scores on each experimental result. Topic modelling was carried out using Latent Dirichlet Allocation (LDA) method which evaluated by its coherence score. The results show that LR is a model that can predict data well in almost all scenarios in this study with the highest score on the Attraction aspect with a score of 84.4%, Amenity 84.2%, Accessibility 89.1%, Image 70%, and HR 92.8%. Meanwhile, DT can predict data well on the Price aspect with a score of 91.3%. From the results of topic modelling, we recommend some approaches for the development of tourism in Borobudur Temple and Prambanan Temple, one of which is the government can lower the price of admission to Prambanan Temple and Borobudur Temple for foreign tourists so that they can compete with tourist attractions in neighboring Indonesia because many reviews state that the price of entrance tickets to Prambanan Temple and Borobudur Temple is too expensive for foreigners.]]>
Copyrights © 2022
