Automatic Analysis Of Descriptive Texts

This paper d e s c r i b e s a sys t em t h a t a t t e m p t s to i n t e r p r e t d e s c r i p t i v e t e x t s w i t h o u t the uJe of complex grammars. The pu rpose of the sys t em i s to t r a n s f o r m the d e s c r i p t i o n s to a s t a n d a r d form which may be used as the b a s i s of a d a t a b a s e s y s tem knowledgeable i n the s u b j e c t m a t t e r of the teXt. The t e x t s c u r r e n t l y used a re w i ld p l a n t d e s c r i p t i o n s taken d i r e c t l y from a p o p u l a r book on the s u b j e c t . P r o p e r t i e s such as s i z e , shape and c o l o u r a r e a b s t r a c t e d f rom the d e s c r i p t i o n s and r e l a t e d to p a r t s of the p l a n t in which we a re i n t e r e s t e d . The r e s u l t i n g o u t p u t i s a s t a n d a r d ined h i e r a r c h i c a l s t r u c t u r e h o l d i n g on ly s i g n i f i c a n t features of the d e s c r i p t i o n . The sys t em, implemented i n t he PROLOG p r o gramming l a n g u a g e , u s e s keywords co i d e n t i f y the way segments of the t e x t r e l a t e to the o b j e c t d e s c r i b e d . I n f o r m a t i o n on words is he ld in a keyword l i s t of nouns r e l a t i n g to p a r t s of the o b j e c t d e s c r i b e d . A d i c t i o n a r y c o n t a i n s the a t t r i b u t e s of o r d i n a r y words used by the sy s t e m to a n a l y s e the t e x t . The t e x t i s d i v i ded i n t o seE" ments u s i n g i n f o r m a t i o n p rov ided by c o n j u n c t i o n s and p u n c t u a t i o n . About half the texts processed are correctly analysed at present. Proposals are made for f u t u r e work to improve this figure. There seems Co be no inherent reason why the technique cannot be generalised so chac any text of seml-standard descriptions can be automatically converted to a canonical form.