Clustering Chinese Regional Cultures with Online-gaming Data

To identify cluster of societies is not easy in subject to the availability of data. In this study, from the prospective of computational social science, we propose a novel method to cluster Chinese regional cultures. Using millions of geotagged online-gaming data of Chinese internet users playing online card and board games with regional features, 336 Chinese cities are grouped into several main clusters. The geographic boundaries of clusters coincide with the boundaries of provincial regions. The north regions in China have more geographical proximity, when regional variations in south regions are more evident.