Research and Development in Natural Language Understanding as Part of the Strategic Computing Program
Annual rept. 21 Dec 1986-20 Dec 1987
BBN SYSTEMS AND TECHNOLOGIES CORP CAMBRIDGE MA
Pagination or Media Count:
In this report two papers stemming from work from 1987 are reprinted. In the first, we report pioneering work in knowledge representation for natural language processing. In BBNs natural language understanding and generation system Janus, we have used a hybrid approach to representation, employing an intensional logic for the representation of the semantics of utterances and a taxonomic language with formal semantics for specification of descriptive constants and axioms relating them. Remarkably, 99.9 of 7,000 vocabulary items in our natural language applications could be adequately axiomatized in the taxonomic language. In the second paper, we overview the problems, issues, and approaches to dealing with ill-formed input. There are several purposes of this paper 1. to define a class of phenomena that may be termed ill-formed input, 2. to differentiate this class from problems of spoken language understanding, 3. to briefly summarize the state of the art in understanding ill-formed input, 4. to indicate areas for further theoretical progress, and 5. to speculate regarding where there may be strong practical pay-off.