ond, the system is robust; redundant encoding of linguistic knowledge is encouraged by the very nature of distributed experts. Also, the notion of timeouts guarantees that comprehension will never be totally blocked by an inadequate or incorrectly designed expert. Third, the model encourages, and provides a framework for, the development of large experts. LIL and SDL seem to be necessary, if not yet sufficient, types of inter- and intraexpert communication primitives. Fourth, the model has the practical advantages of being developmentally modular, and realistic to implement on a large scale. Experts can be written by different researchers, as long as the LIL and SDL conventions are followed. Experts can be stored on disk, and reside in core only on demand from the current sentence.
We are convinced that distributed word experts will prove to be the only acceptable course in modeling human language. In the WEP, we have only scratched the surface of the larger theory. But we hope our ideas are helping to bring that theory into better focus.
We extend our thanks to all members of the AI group for their comments and contributions to this research: Milt Grinberg, Rich Wood, Randy Trigg, Bob Bane, Jim Reggia, Liz Allen, Jim Williams, Mike Konard, Bruce Israel, Bob Krovetz.
The research described herein is funded by NASA, under grant NSG-7253. Their support of this basic research is deeply appreciated.
Bobrow D., & Winograd T. "An overview of KRL, a knowledge representation language". "Cognitive Science", 1977, 1 (No. 1).
Marcus M. "An overview of a theory of syntactic recognition for natural language". "Doctoral dissertation, MIT AI Memo 531", 1979.
Rieger C. "Viewing parsing as word sense discrimination". In W. Dingwall (Ed.), "A survey of linguistic science". Greylock, 1977.
Rieger C. The importance of multiple choice. Proc. TINLAP-2, Urbana, Ill., 1978 (also University of Maryland Computer Science TR-656).
Rieger C., & Small S. Word expert parsing. Proc. IJCAI-29, Tokyo, 1979.
Riesbeck C. Computational understanding: Analysis of sentences and context. Doctoral dissertation, Stanford AI Memo 238, 1974.
Small S. Word expert parsing: A theory of distributed word-based natural language understanding. Doctoral dissertation, University of Maryland computer Science TR-954, 1980.