ABSTRACT
The development of information technology has made it easier for various people activities. One of them is the development of information technology in the tourism sector, where information and access to deliver opinions or reviews on tourist attractions are available on various websites and social media. This study aims to explore the relationships among polarity, subjectivity, and clusters characteristics of online tourist reviews on different tourist objects in the Malang Regency. Online-review data was retrieved using the web content mining technique. The study applies sentiment analysis and k-mean clustering to analyze retrieved online-review data. The clustering process was performed to classify existing reviews based on their respective characteristics using the k-means clustering algorithm. The results show a significant difference between clusters and sentiment of the visitor reviews on each tourist object. The sentiment analysis results also indicate differences in the interpretation of polarity and subjectivity in each cluster at the same objects. The results of each tourist object's clusters and sentiment show that the review of the services and objects offered tends to be a positive sentiment. On the other hand, the reviews for the facilities offered, the visitors' reviews tend to be neutral. For the negative sentiment, it mostly occurs about the price as well as the location of the tourist objects.
