Text Mining on Tourism Destinations in Greater Bandung (Case Study: Tangkuban Perahu and Kawah Putih)
Indonesia has a variety of tourism destinations that attract domestic and foreign tourists. The level of foreign tourist arrivals in 2019 has reached 16 million tourists from various countries. The tourism sector occupies the fourth position as the largest foreign exchange earner for the country, amounting to the US $ 20 billion. West Java is one of the provinces in Indonesia which has a variety of tourism potentials that attract domestic and foreign tourists. One of the three main tourist destinations in West Java is Bandung Raya with nature theme destinations namely Tangkuban Perahu and Kawah Putih. One factor that drives the tourism industry to develop is tourist comfort.
Tourist comfort can be seen from tourist reviews on various travel sites. The development of information communication technology makes a variety of information and data reviews of a tourist spot can be accessed quickly. Review in traveling site can be used as a reference for tourists’ tourism destinations and provide feedback for relevant stakeholders. One of the biggest travel review sites in the world is Trip Advisor.
The purpose of this study is to analyze the reviews on two nature-themed tourism objects namely Tangkuban Perahu and Kawah Putih to obtain valuable information. This method is done by using text mining on all English-language reviews on the Trip Advisor site on tourism object. The steps are to collect data, preprocessing data, matrix term documents. Output consist of term frequency, word cloud and sentiment analysis with emotional classification.
Research findings illustrate that text mining can be applied to travel reviews for tourism destinations in Indonesia. Obtained analysis related to tourist destination ratings above the average that can still be improved, the frequency of tourist terms filled with positive words even though expensive word appear, Ekman's emotional sentiment classification is dominated by emotional joy which means tourists feel comfortable in enjoying the tourism object, and emotional sentiment term frequency of sadness and anger can be stakeholder input for tourism destinations.
Keywords: tourism destination, text mining, review, Bandung, sentiment analysis