ANALYSIS OF TRAVEL REVIEWS ON THE PHILIPPINE TOURISM EXPERIENCE BY WORD2VEC MODEL

Abstract

Travel-related websites are considered an important part of tourism nowadays since they not only provide non-biased reviews of a destination or establishment for potential tourists, they are also used by tourism managers to understand the needs, wants and expectations of the market to further improve their offerings. However, only a number of researches have been conducted with regards to the latter. This research analyzes travelers’ comments and feedback about their Philippine trip to gauge the country’s performance as a tourism destination. A total of 1,717 travel reviews and blogs were gathered from three travel-related websites and analyzed using machine learning techniques such as Sentiment Analysis, Word2Vec Analysis, and Cluster Analysis. The findings revealed that reviews written about the Philippines are mostly positive, but it also points out the tourism components that tourists did not like and must be improved. Furthermore, proposed applications of the Word2Vec output for tourism promotion and development are presented and implications of the research findings in Philippine tourism are discussed.