Online Review Analysis of Tourist Attractions Using Topic Modeling: Focusing on Hallasan Mountain and Seongsan Ilchulbong Peak

Purpose: The purpose of this paper is to examine how the representative tourist attractions of Jeju, Hallasan and Seongsan Ilchulbong, are perceived by tourists. The analysis was conducted by comparing Hallasan Mountain and Seongsan Ilchulbong Peak to more clearly understand the perception of tourists about the two tourist destinations. In particular, in order to exclude the researcher s point of view as much as possible, the online review data of tourism reviews written freely by tourists was used, rather than data collection in a survey method. Research design, data and methodology: In this paper, online review data was collected, frequently appeared words and topics of interest were extracted using text mining techniques, and two tourist sites were compared. By developing a Python web crawler, online reviews of Hallasan and Seongsan Ilchulbong were collected from a world-class tourist website called Tripadvisor. Text pre-processing was performed to remove unnecessary words from online reviews collected through Python program. For each Korean sentence in online review to which text preprocessing was applied, the tokenization technique using the Python library was applied, and morphemes, which are the smallest words with meaning, were extracted. After extracting only words corresponding to nouns, verbs, and adjective parts of speech from morpheme words, the frequency of appearance was calculated. Online reviews converted into morpheme words were classified as common topics through vectorization and topic modeling. In the online review of Hallasan and Seongsan Ilchulbong, frequent words and common themes were compared qualitatively. Conclusions: The most frequent nouns appearing in Hallasan reviews were courses , and Youngsil and Seongpanak appearing frequently, suggesting that tourists prefer the ‘Yeongsil’ and ‘Seongpanak’ courses to the Gwaneumsa course when climbing Hallasan Mountain. The adjective that often appears in Seongsan Ilchulbong was noisy , and Seongsan Ilchulbong was reminiscent of a noisy atmosphere with many tourists. The main topics in the Hallasan Review were Course Information , Seasonal Climbing , and Hiking . The main topics in the Seongsan Ilchulbong review were tourism activities and experience activities . It could be inferred that Hallasan is a tourist destination for tourists, and Seongsan Ilchulbong Peak is recognized as a tourist destination for both experiential and tourist activities.