Tourism Evaluation of Zones for Foreigners Visiting Himeji City Using Support Vector Machines

Abstract

In this paper, a method of evaluating zones from the viewpoint of sight scene resources in Himeji City, Japan, is presented to promote the tourism of that city for foreign tourists. It employs support vector machine (SVM for short). Data presented to discrimination models constructed by SVM learning are prepared from numbers of tourists totaled by country. The element value of the data is therefore equal to the number of the tourists coming from some country and visiting some zone of one square kilometer in Himeji City. The discrimination model judges whether a zone is worth to visit for the tourists coming from each of the following countries: France, United Kingdom, Germany, Spain, Singapore, Australia, and United States of America. Experimental results reveal that the proposed method achieves favorable values of recall, precision, and F-measure on average when training data are prepared from numbers of the tourists coming from six countries out of the above seven ones considered to be of importance for the promotion.