Visualizing Dialectal Variation in Japan through Geo-tagged Tweets
Yuanyuan Wang · 2019
This paper presents a novel linguistic analysis of Twitter for exploring regional differences in dialectal variation across Japan. For this, we proposed a Twitter-based visualization system for dialectal variation by analyzing the differences between locations and dialects in geo-tagged tweets from Japan. A key feature of the proposed system is the ability to infer dialect users' origins in dialectal variation between different dialect users in different regions even when such the dialect users' origins are not explicitly shown in Twitter profiles. In this paper, we discuss our linguistic analysis method of dialects on Twitter from Japan and an evaluation of its effectiveness. In addition, we demonstrate the usefulness of our proposed Twitter-based visualization system for dialectal variation with a questionnaire survey.