When extending a scientific knowledge base with new information, particularly information presented in natural language, it is important that the information be encoded in a form that is compatible with the existing knowledge base. Hand built systems for semantic interpretation and knowledge integration can suffer from brittleness. Methods for learning semantic interpretation and integration exist, but typically require large numbers of aligned training examples. Our approach to semantic integration learns rules mapping from syntactic forms to semantic forms using a knowledge base and a text corpus from the same domain.

PDF

In AAAI Spring Symposium on Semantic Scientific Knowledge Integration 2008.

@inproceedings{aaai08-semantic-paths, title={Improving semantic integration by learning semantic interpretation rules}, author={Michael Glass and Bruce Porter}, booktitle={AAAI Spring Symposium on Semantic Scientific Knowledge Integration}, url="http://www.cs.utexas.edu/users/ai-lab?aaai08-semantic-paths", year={2008} }