Toward Geographic Information Harvesting: Extraction of Spatial Relational Facts from Web Documents

摘要: This paper faces the problem of harvesting geographic information from Web documents, specifically, extracting facts on spatial relations among places. The motivation is twofold. First, researchers Spatial Data Mining often assume that data are already available, thanks to current GIS and positioning technologies. Nevertheless, this not applicable case embedded in without an explicit modeling, such as documents. Second, despite huge amount documents conveying useful information, there much work how harvest these particularly challenging because lack annotated which prevents application supervised learning techniques. In paper, we propose places through unsupervised approach recognizes supposing availability proposed based combined use a ontology prototype-based classifier. A study topological directional reported commented.