The tremendous growth of social media and UserGenerated Content (UGC) can be an opportunity for tourism industry. The purpose of this paper is to use Twitter analysis to explore visitor perceptions of two tourism objects in Indonesia (Borobudur and Prambanan Temple). Using 620 tweets referring to the two biggest temples in Indonesia, this research used text mining and sentiment analysis to find meaningful patterns, popular words, and emotional states in opinions. Twitter users held positive perceptions of Borobudur and Prambanan temple. The number of tweets for Borobudur temple was significantly higher than Prambanan temple. The frequency of tweet from both objects had an upward trend toward the end of the year. This research provides a methodology that future researchers can use in applying social media analytics to explore major issues and extract sentiment information from text messages. Limited research has been conducted applying social media analysis for Indonesia tourism. Therefore, this study fills a gap by using social media analytics with Twitter data to examine the Twitter users’ thoughts and emotions for Indonesia tourism objects (Borobudur and Prambanan temple).

Keywords— Social media analysis, Twitter mining, Indonesia tourism