A Method for Constructing Geographical Knowledge Graph from Multisource Data

Global problems all occur at a particular location on or near the Earth’s surface. Sitting junction of artificial intelligence (AI) and big data, knowledge graphs (KGs) organize, interlink, create semantic knowledge, thus attracting much attention worldwide. Although existing KGs are constructed from internet encyclopedias contain abundant they lack exact coordinates geographical relationships. In light this, graph (GeoKG) construction method based multisource data is proposed, consisting modeling schema layer filling layer. This has two advantages: (1) can be extracted geographic datasets; (2) represented integrated. Firstly, designed to represent knowledge. Then, methods extraction integration fill layer, storage developed associate semantics with geospatial Finally, GeoKG verified through linkage rate, relationship application cases. The experiments indicate that could automatically extract integrate data. Additionally, our higher success rate linking web pages datasets, its have increased 100%. paper bridge distance between Geographic Information System KG, facilitating more applications.