Title:	BRINGING DISCOVERY OF LOW-PROFILE SIGHTSEEING SPOTS FOR INBOUND TOURISTS - ACQUIRING TOURING SERENDIPITY USING A JOINT TOPIC MODEL -
Author(s):	Satona Shirai, Takayasu Yamaguchi, and Hiroshi Uehara
ISBN:	978-989-8704-26-9
Editors:	Piet Kommers and Pedro Isaías
Year:	2021
Edition:	Single
Keywords:	Social Media, Data Mining, Data Analytics, Recommendation, Tourism, Machine Learning
Type:	Short
First Page:	303
Last Page:	308
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	This study proposes to provide foreign tourists with various information concerning sightseeing spots, including unvisited ones, in an interpretable manner according to relevance to their preferences. Recently, enormous amounts of travel reviews posted on world travel websites have become an important resource for travelers. Nevertheless, several sightseeing spots potentially attractive to foreigners are left undiscovered because of the information gap between reviews written by foreigners and those written by native tourists. Although the reviews written by native tourists include a wide variety of information that are unknown to foreigners, the information is not conveyed to foreigners efficiently, resulting in inbound tourism within a limited range of standard spots. In this study, a topic model, latent Dirichlet allocation (LDA), is applied to specify relevant reviews for foreigners from miscellaneous topics in the native reviews. Subsequently, the joint topic model, an extended LDA model, is applied to augmented reviews, including reviews written by foreigners and the native reviews extracted by LDA, and acquires categories of recommendable sightseeing spots as a form of topic. Each topic consists of sightseeing spots and their interpretable information, reflecting the interests of the travelers. The proposal was applied to reviews related to sightseeing spots in Japan, and it successfully acquired topics including a broader range of sightseeing spots than using information based only on foreign reviews. Moreover, several spots rarely known to foreigners were found to comply with their preferences, further supporting the performance of the proposal.
	Go Back