Home // International Journal On Advances in Systems and Measurements, volume 12, numbers 3 and 4, 2019 // View article

Less-Known Tourist Attraction Analysis Using Clustering Geo-tagged Photographs via X-means

Authors:
Jhih-Yu Lin
Shu-Mei Wen
Masaharu Hirota
Tetsuya Araki
Hiroshi Ishikawa

Keywords: Flickr; geo-tagged photograph; less-known tourist attractions；X-means

Abstract:
Today, travelers can readily travel around the world using convenient transportation. Not only are opportunities to go abroad for sightseeing is increasing, but tourism industries of every country are developing indirectly. Moreover, many travelers obtain the latest tourist information from the internet for their journeys. However, most information specifically relates to popular tourist attractions, leading to crowds flocking there, which make tourists feel uncomfortable. Contrary to existing studies, which specifically emphasize analyses of popular tourist attractions, we are striving to disperse crowds from popular tourist attractions and provide more spots for travelers to choose by discovering less-known tourist attractions. This study therefore specifically examines discovery of less-known Japanese tourist attractions under the assumption that these spots exist in unfamiliar cities of tourists. According to results of analyzing geo-tagged photographs on Flickr, we use the X-means algorithm to group Japanese cities into different clusters. X-means is an extension of K-means that improved the shortcomings of K-means and which greatly reduced the probability of being trapped into a local optimum. Furthermore, these clusters were used to survey unfamiliar clusters to Japanese and Taiwanese people. Thereby, we can eliminate spots that are in familiar clusters. We propose a formula for ranking tourist attractions that lets travelers choose these spots easily. Results of verification experiments demonstrated that some less-known tourist attractions appeal to Taiwanese and Japanese. Additionally, we examined some factors that might affect respondents as they decide whether a spot is attractive to them or not.

Pages: 215 to 224

Copyright: Copyright (c) to authors, 2019. Used with permission.

Publication date: December 30, 2019

Published in: journal

ISSN: 1942-261x