Automated OWL Annotation Assisted by a Large Knowledge Base

Widespread adoption of the semantic web depends critically on lowering the "barriers to entry" facing document producers. We describe a system that applies automatic partial parsing of web pages into the representations of the large ResearchCyc ontology, combines this with convenient mixed initiative knowledge capture, and produces an OWL annotated document as output. Semantic web publishers can then use this document as a starting point for more elaborate, manual annotation.