Thematic Linked Data Matching: an Approach based on Geographic Reference Data

Many resources published on the Web of Data are described by either direct or indirect spatial references. These spatial references can be used for data matching or cartographic visualization purposes. Indeed, they may be used as instance matching criterion: two resources that are very close in space may represent the same thing, or at least they may have some semantic relationship. However, heterogeneities between spatial references may make their use as instance matching criterion not very reliable or even impossible. In this article, we propose to reduce the data matching difficulties caused by the heterogeneity of spatial references by means of a geographic reference dataset. We also propose to take advantage of links created between thematic resources and geographic resources for designing better maps for data visualization.