ASKNet: Automatically Generating Semantic Knowledge Networks

The ASKNet project uses a combination of NLP tools and spreading activation to transform natural language text into semantic knowledge networks. Network fragments are generated from input sentences using a parser and semantic analyser, then these fragments are combined using spreading activation based algorithms. The ultimate goal of the project is to create a semantic resource on a scale that has never before been possible. We have already managed to create networks more than twice as large as any comparable resource(1. 5 million nodes, 3. 5 million edges) in less than 3 days. This report provides a summary of the project and its current state of development.

Authors

• Brian Harrington Stephen Clark

Keywords

No keywords are indexed for this paper.