INTERFACILE: Linguistic Coverage and Query Reformulation
Transcription
INTERFACILE: Linguistic Coverage and Query Reformulation
I N T E R F A C I L E Yvette : Linguistic Coverage Mathieu and Paul Query Reformulation Sabatier C N R S - LADL, Universit4 Paris 7 Tour Centrale 9 E 2 Place Jussleu 75005 Paris I. INTRODUCTION - T h e e x p e r i e n c e w e h a v e g a i n e d in d e s i g n i n g [3,6,7] and using natural lanauage i n t e r f a c e s h a s led us to d e v e l o p a g e n e r a l natural language system, INTERFAC[LE, involving the following principles : the linguistic coverage must be elementary but must include phenomena that allow a rapid, concise and s p o n t a n e o u s i n t e r a c t i o n , s u c h as anapho~a (ellipsis, pronouns, etc.) the linguistic competence and limits of t h e i n t e r f a c e m u s t be e a s i l y a n d r a p i d l y p e r c e i v e d by t h e u s e r . t h e i n t e r f a c e m u s t be e q u i p p e d w i t h s t r a t e g i e s a n d p r o c e d u r e s for l e a d i n g t h e u s e r to a d j u s t h i s l i n g u i s t i c c o m p e t e n c e to t h e c a p a c i t i e s of t h e system. a semantlc parser transforms the syntactic analysis into a semantic representation by consulting the knowledge base. - an evaluator consults the knowledge b a s e a n d p r o d u c e s a n s w e r s to q u e s t i o n s given their semantic representation. - a general system guides the user and h e l p s h i m to f o r m u l a t e a n d r e f o r m u l a t e his queries. T h e s y s t e m is e n t l r e l y p r o g r a m m e d in P R O L O G II a n d r u n s on v a r i o u s c o m p u t e r s (VAX, S P S and micro-computers). Responses are given instantaneously. E e r e is a n e x a m p l e INTERFACILE : > comment crder of a dialogue un flchler ? u t i l l s e z la c o m m a n d e CREATE nomflchler : - W e h a v e i l l u s t r a t e d t h e s e p r i n c i p l e s in an application : a natural language (French) i n t e r f a c e for a c q u i r i n g t h e f o r m a l c o m m a n d s of s o m e o p e r a t i n g s y s t e m l a n g u a g e s . ( T h e e x a m p l e s g i v e n h e r e c o n c e r n DCL of D i g i t a l Equipment Corporation). > comment INTERFACILE's The system of son contenu - utilisez l'ddlteur commande : EDIT nomfichler : - a portable syntactic grammar w r i t t e n in t h e f o r m a l i s m of M e t a m o r p h o s i s G r a m m a r s [2]. - modifier dependent of > comment copier rdpertolre ? [4]. a domain dependent knowledge base w r i t t e n in a s u b s e t of p r e d i c a t e l o g i c (Horn Clauses). data and un avec fichier la a syntactic parser coupled with the g r a m m a r a n d t h e l e x i c o n - g r a m m a r (i) p r o d u c e s i n t e r m e d i a t e a n a l y s e s of q u e s t i o n s , a n d (2) g e n e r a t e s s o m e m e s s a g e s in c a s e of e r r o r s . dane un u t l l l s e z la c o m m a n d e COPY nomflchier [nomrdpertoire] : > c o m m e n t r e n o m m e r ce d e r n i e r ? u t i l i s e z la c o m m a n d s : RENAME/DIRECTORY [ancienrdpertoire] [nouveaurdpertoire] - <Dialogue 46 ? French lexicon-grammar O p e r a t i n g on t h e s e l i n g u i s t i c domain knowledge : - : ARCHITECTURE is c o m p o s e d - a domain ? - u t i l i s e z la c o m m a n d e PRINT nom£ichier > comment 2. l'imprimer with I> ( T h e p r o m p t ">" p r e c e d e s t h e u s e r ' s q u e r i e s , t h e p r o m p t ..... p r e c e d e s I N T E R F A C I L E ' s a n s w e r s . E n g l i s h t r a n s l a t i o n s of t h e d i a l o g u e s a r e g i v e n in t h e a p p e n d i x ) . 3. LINGUISTIC COVERAGE 4. INTERFACILE handles only elementary s~ntactlc structures° In order to ensure a concise and spontaneous dialogue, proforsLs have been introduced such as : '° pronouns (le, I ) eur .... -possessives la, les, (son, demonstratlves ce dernier~ sa, en, y, (celui-ci, ) ceux-ci, . .) comment copier rSpe~tol~'e ? un flchier dans a un "- u t l l i s e z la c o m m a n d s : COPY nomfichier [nomrQper'toire] > comment •" ? l ' e f f a c e r '~1 ' ~' renvoi e ~ tapez le : fichier ? (i Pgpertoire ? (2 d e vo'tre c h o i x . num4ro <Dialogue 2> Proform r ~ s o l t t t l o n is b a s e d s o l e l y on lexical a~d syntactic information. This strategy Is v i t a l in t h e f r a m e w o r k of a didact.ic system oriented towards beginners as t h e f o l l o w i n g dialogue shows : > comment envoyer utilisateur ? utilisez HAIL > comment "le" %apez > le un la message commands supprimer renvoie le A un : ? & : message ? (1) utilisateur ? (2) num4ro de votre choix. 2 INTERFACILE n'en salt comment : sait rien, supprimer supprimer supprimer un un un <Dialogue 3> Knowing that error recovery is a c o m p l e x t a s k in t h e " r e s t r i c t e d " framework of programming languages (limited syntax and rJ. g o r o u s l y defined semantics), one can appreciate -the d i f f i c u l t y of e n d o w i n g natural language interfaces with such capabilities. One approach t o "the p r o b l e m of e r r o r recoverJng is t o " a c c e p t " user queries that deviate from the system's expectations and (unbeknown to t h e u s e r ) m a n i p u l a t e them into a form that can be analysed. FoP an overview of t h i s a p p r o a c h , s e e [1]. T h i s method is c o s t l y to i m p l e m e n t , and moreover }%as 'two t h e o r e t i c a l faults : (I) T h e r e is n o g u a r a n t e e that the system's interpretation of a " d e v i a n t " query will correspond to t h e u s e r ' s intention in p o s i n g the question. (2) T h e r e w i l l a l w a y s b e a l i m i t to t h e degree of d e v i a t i o n the system can tolerate before it g i v e s u p t r y i n g to analyse a user query. B u t if t h e t r a n s f o r m a t i o n of d e v i a n t queries into recognizable structures is performed automatically, the user will *lever k n o w w h e r e t h i s l i m i t l l e s , a n d m a y believe that the system c a n m a k e s e n s e of anything he enters. The distinction between the sentences the system interprets correctly i~nd t h o s e w h i c h it e n t i r e l y rejects will seem arbitrary to him. mais envoyer un message 6crire un message fichier r6pertoive message If t h e p r o f o r m resolution had been based on the domain semantics, the system would have produced %he answer for" s u p p r e s s i n g a m e s s a g e . argue that the quality of a n a t u r a l language interface must be estimated not o n l y in t e r m s of i t s l i n g u i s t i c coverage b u t a l s o in t e r m s of t h e p r o c e d u r e s for dealing with incorrect, extragrammatical (i.e. correct but not expected b~ the interface) or s e m a n t i c a l l y deviant sentences. Ue . If a n a m b i g u i t y is d e t e c t e d during the r e s o l u i : i o n of a p r o f o v m , INTERFACILF, asks question~ as i n t h e f o ] . ] o w i n g d i a l o g u e : > REFORHULATION Given the plethora of s y n t a c t i c structures and the complexity of m e n t a l processes involved in n a t u r a l language understanding, the linguistic competence of i n % e f f a c e s yet remains limited. If w e r e s o r t to such systems to r e q u e s t ~ service or obtain information that we lack, we generally know much more than the interface itself on the various w a y s for fo~mulatlng our query. lui, ses .... QUERY The strategy developped and implemented in INTERFACILE is q u i t e d i f f e r e n t . Our philosophy is n o t to m i s l e a d the user concerning the system's real capabilities, but rather to lead his linguistic competence to t h a t of t h e i n t e r f a c e . H e r e is a n e x a m p l e strategy : > comment ajouter d'un flchier ? - illustrating A un pour "ajouter", construction : fichier utilisez such is a contenu la 47 ajouter [quelque [quelque chose] chose] (ou t a p e z / pour & compl4tez > comment ajouter is a n o t h e r example > comment envoyer terminal ? initlallssr initialiser 4> <Dialogue : un m e s s a g e - pour "envoyer", utilisez constructions : envoyer annulet) ... <Dialogue Here c r 4 e r un f i c h i e r ~dlter un fichler effacer un fichler dans The guidance given here goes beyond linguistic reformulation. It c o n c e r n s t h e s e m a n t i c s of t h e a p p l i c a t i o n d o m a i n . N o t e t h a t t h e i n f o r m a t i o n g i v e n to t h e u s e r is n o t p r e p r o g r a m m e d , but is (I) d e d u c e d f r o m t h e k n o w l e d g e b a s e a n d (2) g e n e r a t e d b y t h e linguistic component (grammar, lexicon-grammar and parsers). un les 5. envoyer [quelque [quelque sur > comment 7> [quelque chose] [quelqu'un] & compl4tez un disque une dlsquette (ou t a p e z envoyer chose] chose] / pour annulet) ... CONCLUSION As far as t h e d o m a i n of a p p l i c a t i o n is c o n c e r n e d , I N T E R F A C I L E h a s m u c h in c o m m o n w i t h s y s t e m s l i k e UC [9] or I N T E R I X [5]. T h e o r i g i n a l i t y of o u r effort we have devoted <Dialogue system to : lies in t h e 5> INTERFACILE explains exactly which part t h e q u e s t i o n it d o e s n ' t u n d e r s t a n d : an u n k n o w n w o r d , a v e r b w i t h an i n c o r r e c t p r e p o s i t i o n or u n e x p e c t e d o r d e r of c o m p l e m e n t s , etc. In c a s e o£ s p e l l i n g errors, s o l u t i o n s for c o r r e c t i o n a r e g i v e n to t h e u s e r [8]. of - l i n g u i s t i c p h e n o m e n a s u c h as a n a p h o r a that ensure a natural dialogue. s t r a t e g i e s a n d p r o c e d u r e s for reformulation in o r d e r to l i v e w i t h t h e u n a v o i d a b l e l i n g u i s t i c l i m i t a t i o n s of any interface. To g u i d e t h e u s e r in r e f o r m u l a t i n g h i s query, the grammar and the lexicon-grammar a r e c o n s u l t e ~ as k n o w l e d g e b a s e s . T h e u s e r may reformulate his query with the information given by INTERFACILE. (At a n y m o m e n t , t h e u s e r c a n a l s o o b t a i n l i s t s of v e r b s , n o u n s , p r e p o s i t i o n s , pronouns, etc.). the O u r r e s e a r c h a n d e x p e r i e n c e in d e v e l o p i n g a p p l i c a t i o n s in t h e s e t w o d o m a i n s h a v e led us to b e l i e v e t h a t it is n o t p r a c t i c a l to a d d p r o c e d u r e s to an a l b e a d y e x i s t i n g s y s t e m to m a k e it u s e r - £ r l e n d l y . O n t h e contrary, these problems must be taken into a c c o u n t in t h e d e s i g n p h a s e , as w e h a v e d o n e in d e s i g n i n g I N T E R F A C I L E . 6. A C K N O W L E D G E M E N T S W h e n t h e u s e r h a s a n e r r o n e o u s or i n c o m p l e t e v i e w of t h e a p p l i c a t i o n d o m a i n , I N T E R F A C I L E p o i n t s out h i s m i s u n d e r s t a n d i n g , as in t h e following dialogue : > comment version imprimer ? le flchler de quoi p l u t f t de : fichier fichier version fichler ? INTERFACILE example > comment We thank the Centre Nondial d'Informatique (CNI, P a r i s ) for g i v i n g us t h e o p p o r t u n i t y to d e v e l o p a v e r s i o n of I N T E R F A C I L E on a V A X 785. 7. REFERENCES [i] C a r b o n e l l J., H a y e s P., " R e c o v e r y S t r a t e g i e s for P a r s i n g E x t r a g r a m m a t i c a l L a n g u a g e " , A m e r i c a n J o u r n a l of C o m p u t a t i o n a l L i n g u i s t i c s , 9, 3-4, 1983. 6> : initialiser INTERFACILE n'en salt comment : 48 parle (tout simplement) d'un r~pertolre d'un fichier <Dialogue Another d'une W e a r e gra'te£ul to H a n n a h B l a u for h e r f r u i t f u l c o m m e n t s on a p r e l i m i n a r y v e r s i o n of t h i s p a p e r . un fichier salt rien, ? mais [2] C o l m e r a u e r A., Les G r a m m a i r e s d e N4tamorphoses, GIA, L u m i n y , U n i v e r s i t 4 Alx-Narseille If, 1975, a n d " M e t a m o r p h o s i s G r a m m a r s " in N a t u r a l L a n g u a g e C o m m u n i c a t i o n W i t h C o m p u t e r s , B o l c L. ed., S p D l n g e r V e r l a g 1978. [3] D u c h i e r D., S a b a t i e r P., I N T E R F R A N C E : Un prototype d'interface en l a n g u e franGalse, Rapport de recherche, CGE, Harcoussis, 1982. "it" - refers enter and the P r o c . of [4] G r o s s H . , " L e x l c o n - G r a m m a r s syntactic analysis of F r e n c h " , COLING, Karlsruhe, 1984. Dialogue 3 > how [5] G u e z S . , S a b b a g h S., "INTERIX: Un systbme d'aide & l'utlllsation d'Unix", Confdrence AFCET, Paris, 1984. [6] O l l v e l r a E., P e r e l r a L., S a b a t i e r P., "ORBI: An expert system for environmental resources evaluation through natural language, P r o c . of F i r s t International Logic Programming Conference, Marsellle, 1982. > how [7] P i q u e J . F . , S a b a t i e r P., " A n i n f o r m a t i v e , adaptable and efficient natural language consultable database system", P r o c . of E C A I Conference, 1982. >2 the I send use the HAIL can - [8] S a b a t l e r P., A P r o l o g spelling correction, Note 1985. : Translation Dialogue i command "it" refers the of sample I create can > how a file command filename I print I it modify > how can I copy directory ? Dialogue 4 dialogues can > how ? (I) (2) choice. know, but he : a can 5 > how to a file the content contents by t h e Copy the a into a Dialogue latter can 6 I delete can [olddirectory] [newdirectory] into a directory? ? a message in use the [something] [something] to on (or type I send cancel) 7 : [somebody] [something] / to cancel) I prin't a file of a version ? INTERFACILE ? "talks a directory of a f i l e : can - terminal? structures a file a f i l e of a version Dialogue a ... file of what about : : file it / to : ? use the command : COPY filename [directoryname] can type ... "send", complete > how file I send send send ? > how I I append for the command : filename [dlrectoryname] I rename (or : can : : can - message user ? of y o u r ? ? its - use the command RENAME/DIRECTORY > how : for "append" use the structure : append [something] to [something] > how use COPY - 2 to ? send a message write a message suppress a file suppress a directory suppress a message Dialogue - use the editor, command : EDIT filename Dialogue ? : use the command PRINT fllename can > how it INTERFACILE doesn't k n o w s h o w to : > how use the CREATE - a : number complete > how to : can - user > how can I append of a f i l e ? APPENDIX > how a message program for technique, LADL, R., " T a l k i n g to U N I X in E n g l i s h : of U C " P r o c . of A A A I C o n f e r e n c e , [9] U i l e n s k y An overview 1982. (i) (2) choice. I suppress enter - : : can - to file ? directory ? number of y o u r I initialize a file ? INTERFACILE doesn't know, but he knows how to : create a file edit a file delete a file initialize a floppy disk initialize a disk 49