Natural Language Access To A Large Data Base: An Engineering Approach

A n i n t e l l i g e n t program which accepts n a t u r a l language q u e r i e s can a l l o w a n o n t e c h n i c a l user to e a s i l y o b t a i n i n f o r m a t i o n from a l a r g e non-uniform d a t a base. T h i s paper discusses the d e s i g n of a program which w i l l t o l e r a t e a wide v a r i e t y o f r e quests i n c l u d i n g ones w i t h pronouns and r e f e r e n t i a l phrases. The system embodies a c e r t a i n amount o f common sense, so t h a t f o r example, i t "knows when i t does or does not understand a p a r t i c u l a r request and i t can bypass a c t u a l d a t a base search in answering unreasonable r e q u e s t s . The system is c o n c e p t u a l l y simple and could be e a s i l y adapted to o t h e r data bases. 1. I n t r o d u c t i o n The prime o b s t a c l e f o r n o n t e c h n i c a l people who wish to use computers is the need to l e a r n a s p e c i a l language f o r communicating w i t h the mac h i n e . W e f e e l t h a t the time i s r i p e f o r n a t u r a l language systems which can be used by persons who are not t r a i n e d i n any s p e c i a l computer language. Such a system must embody a degree of "common sense," must have a r e l a t i v e l y l a r g e and complete v o c a b u l a r y f o r the s u b j e c t m a t t e r t o b e t r e a t e d , must accept a wide range of grammatical c o n s t r u c t i o n s , and of course must be capable of p r o v i d i n g the i n f o r m a t i o n and computations requested by the user. In o r d e r to design such an a m b i t i o u s system, i t i s i m p e r a t i v e t h a t the u n i v e r s e o f d i s c o u r s e b e l i m i t e d i n some way. Th i s paper d e s c r i b e s work done on a n a t u r a l language q u e s t i o n a n s w e r i n g system f o r a data base c o n t a i n i n g d e t a i l e d records of U.S. Naval a i r c r a f t maintenance and f l i g h t i n f o r mation over a p e r i o d of time. While the s u b j e c t m a t t e r o f t h i s system i s t h e r e f o r e q u i t e c o n s t r a i n ed, we f e e l t h a t the issues we are c o n f r o n t i n g are nonetheless g e n e r a l . I n p a r t i c u l a r , w e w i l l des c r i b e the r e p r e s e n t a t i o n o f common sense i n f o r m a t i o n and procedures, problems of pronoun r e f e r e n c e and storage o f p a r t i a l r e s u l t s , the a b i l i t y t o answer vague o r p o o r l y d e f i n e d q u e s t i o n s , the a b i l i t y of the system to r e c o g n i z e when requests are meaningless or too p o o r l y f o r m u l a t e d to answer, as w e l l a s l i n g u i s t i c issues.