Learning the Space of Word Meanings for Information Retrieval Systems

6. C o n c l u s i o n W e p r o p o s e d to r e p r e s e n t t h e m e a n i n g s of w o r d s in space . W e g a v e a l e a r n i n g m e c h a n i s m fo r t i le space . W e p r o v e d ti le e f f e c t i v e n e s s of t h e p r o p o s e d m e t h o d s in an i n f o r m a t i o n r e t r i e v a l s y s t e m . O n e of t h e r e a s o n s we c h o s e t h e f ie ld of J a p a n e s e l i t e r a t u r e was t h a t t h e r e was s t r o n g d e m m l d f r o m th.e r e s e a r c h e r s in t h a t f ie ld to m a k e a u s e f u l s y s t e m ; s y s t e m s b a s e d on c o n v e n t i o n a l k e y w o r d s d id r iot w o r k welt f o r the f ie ld of l i t e r a t u r e . A n o t h e r r e a s o n is t h a t t h e f i e ld o f f e r s g o o d e x a m p l e s of e n t i t i e s fo r w h i c h u n i v e r s a l s y m b o l i c r e p r e s e n t a t i o n is diff icul t . To app ly o u r m e t h o d to o t h e r f i e lds , we m u s t s o l v e o n l y two p r o b l e m s . O n e p r o b l e m is to s e l e c t s y m b o l s to be p u t ill t i le in i t ia l space . In l i t e r a t u r e , w o r k s a n d w r i t e r s p l a y e d t i le ro le . Ti~e o t h e r p r o b l e m is to d e t e r m i n e c r i t e r i a f o r space c o n f i g u r a t i o n . W e u s e d a u s e r ' s s a t i s f a c t i o n wittl t h e a n s w e r s f r o m th e i n f o r m a t i o n r e t r i e v a l s y s t e m as ti le c r i t e r i o n . If t h e s e two p r o b l e m s are s o l v e d , o u r m e t h o d c a n be a p p l i e d to a n y d o m a i n . " l 'hcsc p r o b l e m s do u o t s e e m so diff icul t . F r o m the v i e w p o i n t of ar t i f ic ial i n t e l l i g e n c e , the. s e m a n t i c space Rives a ba s i s f o r s t u d i e s on a b d u e t i o n an.d a n a l o g y . T h e d i s e o v e r y of b l a n k s p a c e snr.r o u n d e d by s y m b o l s carl l ead to d i s c o v e r i n g of n e w idea s by m a c ] t i n e s . S iuce we c a n m e a s u r e s e m a n t i c s i m i l a r i t y d i r ec t ly by d i s t a n c e in s e m a n t i c s p a c e , wc can m a k e a n a l o g i c a l r e a s o n i n g b a s e d oil t h a t s imilar. ity. F o r t h e s a m e r e a s o n , f r o m t h e l i n g u i s t i c p o i n t of v i ew , t h e s e m a n t i c spac, e e ~ l be u s e f u l f o r under . . s t a n d i n g m e t a p h o r i c a l e x p r e s s i o n s u c h as " H o f s . . t a d t e r is P r o l o g b e t w e e n p r o c e d u r a l s e m a n t i c s a n d d e c l a r a t i v e s e m a n t i c s " . R e f e r e n c e s [llAllen,,L:Recognizing Intentions from Natural Language Utterances, in Brady,M,,Berwiek,R.C (eds.): CoTnputatlonal Models of Di~eou'me. MIT Press(1983) 121Barwiz e, J., Perry. J. : Situaiians and A ltit, udes, M 1T Press(1983), [3JDowty,D.R, et al,: Introdw:tion to Monta que Sranantics, D. Rid el P ublishing ( 19111 ) [41Feldman,J.A,:Conneetionist Models and Their Applications Introduction, Coynilive So.brace, Vol.9,pp. 1-3(1985). [5]Kaplan,J.:Cooperatiw. • Responses from a Portable Natural Language Database Qury System, in Brady,M.,Berwick,R.C.(eds): Computational Models o f Discourse, MIT Press(1983). [6]LebowiLz,M.:Categorizing Numeric Information for' General.. ization, Coffnlfive Science, Vol, 9,pp.285-308 (1985). [?]Rieger,B,B,:Procedural Meaning Representation by Connotative Dependency Struetures. An Empirical Approach to Word Semantics for Analogical Infereneing, Prec. COLm c . a 2 (1982) [8]Rieger,BB,:SemaaLie Relevance and Aspect Dependency in a Given Subject i)omain;