A data-based approach to identifying regional typologies and exemplars across the urban–rural gradient in Europe using affinity propagation
运用亲和传播算法，无预设阈值识别欧盟区域类型，并基于气候、土地覆盖等变量找出各类典型区域，揭示城乡差异的自然与人文驱动因素。
We apply recent developments in data-mining and statistics, using affinity propagation (AP) to identify regional typologies in the European Union (EU) and characterize major factors between rural–rural and rural–urban regional differences, without predetermined thresholds. We identify a representative ‘exemplar’ within each cluster using the drivers of Copus enriched with climate and land-cover/land-use variables to provide geographical context and pinpoint differences driven by natural and human–natural landscapes. Building upon the works of Dijkstra and the Eudora Project, we expand the dimensions of regional differences, introducing a threshold-less, data-driven model able to identify exemplars, and the main characteristics of each cluster or regional typology.