Authors: Dalia Varanka*, United States Geological Survey
Topics: Geographic Information Science and Systems
Keywords: ontology alignment, semantic interoperability, geospatial semantics
Session Type: Virtual Paper
Start / End Time: 11:10 AM / 12:25 PM
Room: Virtual 46
Presentation File: No File Uploaded
Knowledge Graphs (KG) are virtual data layers to search and extract information from a variety of sources to an applied ontology schema for reasoning. Ontologies are combinations of scientifically and socially constructed artifacts. Different perspectives can be matched with predominant models through alignment. The process of matching ontologies clarifies the types of knowledge users practice and require, as well as whether data will be interoperable. The interoperability framework for a prototype geospatial KG called MapKB developed at the U.S. Geological Survey focuses on processes of semantic articulation, data sharing and context, and semantic alignment. Semantic articulation involves locally specific information by which users draw on their own resourcefulness to express knowledge compressed as semiotic information. Data sharing is a coordinated group process that requires accountability and trust. Data context is the ability for a system to efficiently and effectively respond to queries. Semantic alignment is a cognitive model of mutual understanding. Phases in the technical development of the framework include: feature instance and metadata integration within graphs; formal publication as Semantic Web linked data and schemas; and ontology alignment. WorldView is a mapping product for geospatial ontology alignment. Surface water domain schemas of the U.S. National Hydrography Dataset, Cree hydrography ontology as reported by Wellen and Sieber (2013), and native Spanish natural language. The interface includes a comparison of feature labels, subgraph semantics, and Boolean diagrams to visualize relations. The geometric feature footprints appear as a map.