Steps Toward the Alignment of Complementary Lexical Resources and Knowledge Databases
Abstract
Since 1997, the FrameNet project at the International Computer Science Institute has been developing a uniquely detailed lexicon of English based on Frame Semantics and manually annotated examples from a balanced corpus, and has distributed copies of the lexicon and the annotations to a wide variety of researchers in natural language processing. This contract funded a meeting to evaluate the status of the project and suggest ways in which it could be more useful to the government and other clients. Based on the suggestions of the evaluators, the FrameNet team used the opportunity to increase the coverage of the lexicon, to rapidly add new frame relations and semantic types, to develop software and techniques for full-text annotation, to improve the public website, documentation, data consistency and data distribution system, and to plan for more a more automated workflow and closer connections to WordNet and other knowledge resources, depending on future funding and collaboration with related projects around the world.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2006
- Accession Number
- ADA444649
Entities
People
- Charles J. Fillmore
- Collin F. Baker
Organizations
- International Computer Science Institute