7/27/2019 P98-2196

1/7

R e c o g n i z i n g S y n t a c t i c E r r o r s i n t h eW r i t in g o f S e c o n d L a n g u a g e L e a rn e rs *D a v i d S c hne i de r a nd K a t h l e e n F . M c C oyDepa rtme nt of Linguistics Comp ute r and Informati on SciencesUniversi ty of Delaware Universi ty of DelawareNewark, DE 19716 Newark, DE 19716{dschneid,mccoy}@cis.udel.edu

A bs t r a c tThis paper reports on the recognition compo-nent of an intelligent tutoring system that isdesigned to help foreign language speakers learnstandard English. The system models the gram-mar of the learner, with this instantiation ofthe system tailored to signers of American SignLanguage (ASL). We discuss the theoret ical mo-tivations for the system, various difficulties thathave been encountered in the implementation,as well as the methods we have used to over-come these problems. Our method of cap-turing ungrammaticalities involves using mal-rules (also called 'error productions'). However,the straightforward addition of some malrulescauses significant performance problems withthe parser. For instance, the ASL populationhas a strong tendency to drop pronouns and theauxilia ry verb 'to be'. Being able to accountfor these as sentences results in an explosionin the number of possible parses for each sen-tence. This explosion, left unchecked, greatlyhampers the performance of the system. Wediscuss how this is handled by taking into ac-count expectations from the specific population(some of which are capture d in our unique usermodel). The different representations of lexicalitems at various points in the acquisition pro-cess are modeled by using mal-rules, which ob-viates the need for multiple lexicons. The gram-mar is evaluated on its ability to correctly di-agnose agreement problems in actual sentencesproduced by ASL native speakers.1 Overv iewThis paper reports on the error-recognitioncomponent of the ICICLE (Interactive Com-puter Identification and Correction of LanguageErrors) system. The system is designed to bea tutorial system for helping second-language(L2) learners of English. In this instant iation" T h i s w o r k w a s s u p p o r t e d b y N S F G r a n t# S R S 9 4 1 6 9 1 6 .

of the system, we are focusing on the par-ticular problems of American Sign Language(ASL) native signers. The system recognizeserrors by using mal-rules (also called 'error-product ion rules') (Sleeman, 1982), (Weischedelet al., 1978) which exte nd the language acceptedby the grammar to include sentences contain-ing the specified errors. The mal-rules them-selves are derived from an error t axonomy whichwas the result of an analysis of writing samples.This pape r focuses primarily on the unique chal-lenges posed by developing a grammar that al-lows the pa rser to efficiently parse and recog-nize errors in sentences even when multiple er-rors occur. Additionally, it is importan t to notethat the users will not be at a uniform stageof acquisition - the system must be capable ofprocessing the input of users with varying lev-els of English competence. We briefly describehow acquisition is modeled and how this modelcan help with some of the problems faced by asystem designed to recognize errors.

We will begin with an overview of the entireICICLE system. To motivate some of the dif-ficulties encountered by our mal-rule-based er-ror recognition system, we will briefly describesome of the errors common to the populationunder study. A major problem that must befaced is parsing efficiency caused by mul tipleparses. This is a particular ly difficult problemwhen expected errors include omission errors,and thus this class of errors will be discussedin some detail. Another important problem in-volves the addition/subtraction of various syn-tactic features in the g ramma r and lexicon dur-ing acquisition. We describe how our systemmodels this without the use of multiple lexicons.We follow this by a description of the currentimplementation and grammar coverage of thesystem. Finally, we will present an evaluat ionof the system for number/agreement errors inthe target group of language learners.

1198

7/27/2019 P98-2196

2/7

2 S y s t e m O v e r v i e wT h e I C I C L E s y s t e m i s m e a n t t o h e l p s e c o n d -l a n g u a g e l e a r n e r s b y i d e n t i f y i n g e r r o r s a n d e n -g a g i n g t h e l e a r n e r s in a t u t o r i a l d i a l o g u e . I tt a k e s a s i n p u t a t e x t w r i t t e n b y t h e s t u d e n t .T h i s i s g i v e n t o t h e e r r o r i d e n t i f i c a t i o n c o m p o -n e n t , w h i c h i s r e s p o n s i b l e f o r f l a g g i n g t h e e r -r o r s . T h e i d e n t i f i c a t i o n i s d o n e b y p a r s i n g t h ei n p u t o n e s e n t e n c e a t a t i m e u s i n g a b o t t o m -u p c h a r t p a r s e r w h i c h i s a s u c c e s s o r t o ( A l l e n ,1 9 95 ). T h e g r a m m a r f o r m a l i s m u s e d b y t h ep a r s e r c o n s i s t s o f c o n t e x t - f r e e r u l e s a u g m e n t e dw i t h f e a t u r es . T h e g r a m m a r i ts e l f is a g r a m -m a r o f E n g l i sh w h i c h h a s b e e n a u g m e n t e d w i t ha s e t o f m a l - r u le s w h i c h c a p t u r e e r r o r s c o m m o nt o t h i s u s e r p o p u l a t i o n . W e w i l l b r i e f l y d i s c u s ss o m e c l a s s es o f e r r o r s t h a t w e r e u n c o v e r e d i no u r w r i t i n g s a m p l e a n a l y s i s w h i c h w a s u s e d t oi d e n t i f y e r r o r s e x p e c t e d i n t h i s p o p u l a t i o n . T h i sd i s c u s s i o n w i ll m o t i v a t e s o m e o f t h e m a l - r u l e sw h i c h w e r e w r i t t e n t o c a p t u r e s o m e c l a s s e s o fe r r o r s , a n d t h e d i f f i c u l ti e s e n c o u n t e r e d i n i m -p l e m e n t i n g t h e s e m a l -r u l e s. T h e m a l - r u l e s a r es p e c i a l l y t a g g e d w i t h i n f o r m a t i o n h e l p f u l i n t h ec o r r e c t i o n p h a s e o f t h e s y s t e m .T h e e r r o r i d e n t i f i c a t i o n c o m p o n e n t r e l i e s o ni n f o r m a t i o n i n t h e u s e r m o d e l - t h e m o s t i n t e r-e s t i n g a s p e c t o f w h i c h i s a m o d e l o f t h e a c q u i si -t i o n o f a s e c o n d l a n g u a g e . T h i s m o d e l ( i n s ta n -t i a t e d w i t h i n f o r m a t i o n f r o m t h e A S L / E n g l i s hl a n g u a g e m o d e l ) i s u s e d t o h i g h l i g h t t h o s eg r a m m a r r u l es w h i c h th e s t u d e n t h a s m o s t l ik e lya l r e a d y a c q u i r e d o r i s c u r r e n t l y i n t h e p r o c e s so f a c q u i r i n g . T h e s e r u l e s w i l l b e t h e o n e s t h ep a r s e r a t t e m p t s t o u s e w h e n p a r s in g t h e u s e r ' si n p u t . T h u s w e t a k e a n i n t e r la n g u a g e v ie w o ft h e a c q u i s i t i o n p r o c e s s ( S e l i n k e r , 1 9 7 2 ), ( E l l i s ,1 9 9 4) , ( C o o k , 1 99 3) a n d a t t e m p t t o m o d e l h o wt h e s t u d e n t ' s g r a m m a r i s li k el y t o c h a n g e o v e rt i m e . T h e e s s e n c e o f t h e a c q u i s i ti o n m o d e l ist h a t t h e r e a r e d i s c r e t e s t a g e s t h a t a l l l e a r n e r s o fa p a r t i c u l a r l a n g u a g e w i ll g o t h r o u g h ( K r a s h e n ,1 9 81 ) , ( I n g r a m , 1 9 89 ) , ( D u l a y a n d B u r t , 1 9 7 4 ),( B a i l e y e t a l . , 1 9 7 4 ). Ea c h o f t h e s e s t a g e s i sc h a r a c t e r i z e d i n o u r m o d e l b y s e t s o f la n g u a g ef e a t u r e s ( a n d t h e r e f o r e c o n s t r u c t i o n s ) t h a t t h el e a r n e r i s i n t h e p r o c e s s o f a c q u i r in g . I t i s a n t ic i -p a t e d t h a t m o s t o f t h e e r r o r s t h a t l e a rn e r s m a k ew i ll b e w i t h i n t h e c o n s t r u c t i o n s ( w h e r e " c on -s t r u c t io n " is c o n s t r u e d b r o a d l y ) t h a t t h e y a r e i nt h e p r o c e s s o f a c q u i r i n g ( V y g o t s k y , 1 98 6) a n dt h a t t h e y w i l l f a v o r s e n t e n c e s i n v o l v i n g t h o s ec o n s t r u c t i o n s i n a " h y p o t h e s i z e a n d t e s t" s t y leo f le a r n in g , a s p r e d i c t e d b y i n t e r l a n g u a g e t h e -o r y. T h u s , t h e p a r s e r f a v o rs g r a m m a r r u l e s in -v o l v i n g c o n s t r u c t i o n s c u r r e n t l y b e i n g a c q u i r e d( a n d , t o a l es s e r e x t e n t , c o n s t r u c t i o n s a l r e a d y

a c q u i r e d ) .T h e c o r r e c t io n p h a s e o f t h e s y s t e m i s a f o c u so f c u r r e n t r e s ea r c h . A d e s c r i p t i o n o f t h e s t r a te -g i e s f o r t h i s p h a s e c a n b e f o u n d i n ( M i c h a u da n d M c C o y , 1 9 9 8 ) a n d ( M i c h a u d , 1 9 9 8 ) .3 E x p e c t e d E r r o r sI n o r d e r t o i d e n t i f y t h e e r r o r s w e e x p e c t t h ep o p u l a t i o n t o m a k e , w e c o l l e c t e d w r i t i n g s a m -p l e s f r o m a n u m b e r o f d i f f e r e n t s c h o o l s a n d o r -g a n i z a t i o n s f o r t h e d e a f . T o h e l p i d e n t i f y a n yi n s t a n c e s o f l a n g u a g e t r a n s f e r b e t w e e n A S L a n dw r i t t e n E n g l i s h , w e c o n c e n t r a t e d o n e l i c i t i n gs a m p l e s f r o m d e a f p eo p l e w h o a r e n a t i v e A S Ls i g n e r s . I t i s i m p o r t a n t t o n o t e t h a t A S L i s n o ts i m p l y a t ra n s l a t i o n o f s t a n d a r d E n g l i s h i n tom a n u a l g e s t u re s , b u t r a t h e r is a c o m p l e t e l an -g u a g e w i t h i t s o w n s y n t a x , w h i c h i s s i g n i f i c a n tl yd i f f e r e n t f r o m E n g l i sh . S o m e o f o u r p r e v i o u sw o r k ( S u r i a n d M c C o y , 1 9 9 3) e x p l o r e d h o w l a n -g u a g e t r a n s f e r m i g h t i n f lu e n c e w r i t t e n E n g l i s ha n d s u g g e s t e d t h a t n e g a t i v e l a n g u a g e t r a n s f e rm i g h t o c c u r w h e n t h e r e a l i z a t i o n o f s p e ci f ic la n -g u a g e f e a t u r e s d i f f e r e d b e t w e e n t h e f i r s t la n -g u a g e a n d w r i t t e n E n g li s h . F o r in s t a n c e , o n ef e a t u r e i s t h e r e a l i z a t i o n o f t h e c o p u l a " b e " . I nA S L t h e c o p u l a " b e " i s o f t e n n o t l e x i c a l i z e d .T h u s , n e g a t iv e l a n g u a g e t r a n s f e r m i g h t p r e d i c to m i s s i o n e r r o r s r e s u l t i n g f r o m n o t l e x ic a l i zi n gt h e c o p u l a " b e " i n t h e w r i t t e n E n g l i sh o f A S Ls ig n e rs . W h i l e w e c o n c e n t r a t e h e r e o n e rr o r sf r o m t h e A S L p o p u l a t i o n , t h e e r r o r s i d e n ti f ie da r e l i k e ly to b e f o u n d i n l e a r n e r s c o m i n g f r o mf i r st l a n g u a g e s o t h e r t h a n A S L a s w e l l. T h i sw o u l d b e t h e c a s e i f t h e f i r s t l a n g u a g e h a s f e a -t u r e s i n c o m m o n w i t h A S L . F o r i n s t a n c e th em i s s i n g c o p u l a " b e " i s a l s o a c o m m o n e r r o r i nt h e w r i t i n g o f n a t i v e C h i n e s e s p e a k e r s s i nc e C h i -n e s e a n d A S L s h a r e t h e f e a t u r e t h a t t h e c o p u l a" b e " i s o f t e n n o t l e x i c a li z e d . T h u s , t h e e x a m -p l e s s e e n h e r e w i l l g e n e r a l i z e t o o t h e r l a n g u a g e s .I n t h e f o l lo w i n g w e d e s c r i b e s o m e c l a ss e s o fe r r o r s w h i c h w e u n c o v e r e d ( a n d a t t e m p t t o " e x-p l a in " w h y a n A S L n a t i v e m i g h t c o m e to m a k et h e s e e r r o r s ) .3 .1 C o n s t i t u e n t O m i s s i o n sL e a r n e r s o f E n g li s h a s a s e c o n d l a n g u a g e ( E S L )o m i t c o n s t i t u e n t s f o r a v a r i e t y o f r e a s o n s . O n ee r r o r t h a t i s c o m m o n f o r m a n y A S L l e a r n e r s ist h e d r o p p i n g o f d e t e r m i n e r s . P e r h a p s b e c a u s eA S L d o e s n o t h a v e a d e t e r m i n e r s y s t e m s im i -l a r t o t h a t o f E n g l i s h , i t is n o t u n u s u a l f o r ad e t e r m i n e r t o b e o m i t t e d a s i n :

( 1 ) I a m _ t r a n s f e r s t u d e n t f r o m . . . .T h e s e e r r o r s c a n b e f l a g g e d r e a s o n a b l y w e l lw h e n t h e y a r e s y n t a c t i c ( a n d n o t p r a g m a t i c ) i n

1199

7/27/2019 P98-2196

3/7

n a t u r e a n d d o n o t p o s e m u c h a d d i t i o n a l b u r d e no n t h e p a r s e r / g r a m m a r .H o w e v e r , m i s s i n g m a i n v e r b s ( m o s t c o m -m o n l y m i s s in g c o p u l a s ) a r e a l so c o m m o n i n o u rw r i t i n g s a m p l e s :( 2 ) O n c e t h e s i t u a t i o n c h a n g e s t h e y _ d i f f e r e n tpeople.

O n e e x p l a n a t i o n f o r t h i s ( a s w e l l a s o t h e rm i s s i n g e l e m e n t s s u c h a s m i s s i n g p r e p o s i t i o n s )i s t h a t c o p u l a s a r e n o t o v e r t l y l e x i c a l i z e d i nA S L b e c a u s e t h e c o p u l a ( p r e p o s i t io n ) i s g o t -t e n a c r o s s i n d i f f e r e n t w a y s i n A S L . B e c a u s e t h ec o p u l a ( p r e p o s i t i o n ) i s r e a l i z e d i n a r a d i c a l l y d i f -f e r en t f a s h i o n i n A S L , t h e r e c a n b e n o p o s i t i v el a n g u a g e t r a n s f e r f o r t h e s e c o n s t r u c t i o n s .I n a d d i t i o n t o o m i t t i n g v e r b s , so m e N P s m a ya l s o b e o m i t t e d . I t h a s b e e n a r g u e d ( s ee , f o re x a m p l e ( L i l l o - M a r t i n , 1 9 9 1 ) ) t h a t A S L a ll o w st o p i c N P d e l e t i o n ( H u a n g , 1 9 84 ) w h i c h m e a n st h a t t o p i c n o u n p h r a s e s t h a t a r e p r o m i n e n t i nt h e d i s c o u r s e c o n t e x t m a y b e l e ft o u t o f a se n -t e n c e. C a r r y i n g t h i s s t r a t e g y o v e r t o E n g l i s hm i g h t e x p l a in w h y s o m e N P s a r e o m i t t e d f r oms e n t e n c e s s u c h a s :( 3 ) Wh i le l i v ing a t co l lege I spend lo t o f m on eybecause _ go ou t t o ea t a lmos t eve ryday .

M a l - r u l e s w r i t t e n t o h a n d l e t h e s e e r r o r s m u s tc a p t u r e m i s s i n g v e r b s , N P s , a n d p r e p o s i t i o n s .T h e g r a m m a r is f u r th e r c o m p l i ca t e d b e c a u s eA S L n a t i v e s a l s o h a v e m a n y e r r o r s i n r e l a t i v ec l a u s e f o r m a t i o n i n c l u d i n g m i s si n g r e l a ti v e p r o -n o u n s . T h e p o s s i b i l i t y o f a ll o f t h e s e o m i s s io n sc a u s e s t h e p a r s e r t o e x p l o r e a g r e a t n u m b e r o fp a r s e s ( m a n y o f w h i c h w i l l c o m p l e t e s u c c e s s-f u l l y ) .3 .2 H a n d l i n g O m i s s i o n sA s w e j u s t s a w , o m i s s i o n s a r e f r e q u e n t i n t h ew r i t i n g o f A S L n a t i v e s a n d t h e y a r e d i f fi c u lt t od e t e c t u s i n g t h e m a l - r u l e f o r m a l is m . T o c l e ar l ys e e th e p r o b l e m , c o n s i d e r t h e f o ll o w i n g t w o s e n -t e n c e s , w h i c h w o u l d n o t b e u n u s u a l i n th e w r i t -i n g o f a n A S L n a t i ve .

( 4 ) The boy happy .( 5 ) I s h a p p y .A s t h e r e a d e r c a n s e e , i n ( 4 ) t h e m a i n v e r b" b e " i s o m i t t e d , w h i l e t h e s u b j e c t i s m i s s i n g i n(5) .T o h a n d l e t h e s e t y p e s o f s en t e n c e s , w e i n -c l u d e d i n o u r g r a m m a r m a l - r u l e s l i k e t h e f o l -l o w i n g :( 6 ) V P ( e r r o r + ) - + A d j P(7) S(error +) -+ VP

A s i g n i f i c a n t p r o b l e m t h a t a r i s e s f r o m t h e s er u l e s is t h a t a s i m p l e a d j e c t i v e i s p a r s e d a s a n Se v e n i f i t i s i n a n o r m a l , g r a m m a t i c a l s e n t e n c e .T h i s b e h a v i o r l e a d s t o m a n y e x t r a p a r s e s , s i n c et h e S w il l b e a b l e t o p a r t i c i p a t e i n l o ts o f o t h e rp a r s e s. T h e p r o b l e m b e c o m e s m u c h m o r e se ri -o u s w h e n t h e o t h e r p o s s i b l e o m i s s i o n s a r e a d d e di n t o t h e g r a m m a r . H o w e v e r , c l o se r e x a m i n a t i o no f o u r w r i t in g s a m p l e s i n d i c a t e s t h a t , e x c e p tf o r d e t e r m i n e r s , o u r u s e r s g e n e r a l l y l e av e o u ta t m o s t o n e w o r d ( c o n s t i t u e n t ) p e r s e n te n c e .T h u s i t i s u n l i k e ly t h a t " h a p p y " w i ll e v e r b e a ne n t i r e s e n t e n c e . W e w o u l d l ik e t h i s f a c t t o b er e f le c t e d i n th e a n a l y s e s e x p l o r e d b y t h e p a r s e r.H o w e v e r , a t r a d i t i o n a l b o t t o m - u p c o n t e x t - f r e ep a r s e r h a s n o w a y t o d e a l w i t h t h i s c a s e , a s t h e r ei s n o w a y t o b l o c k r u l e s f r o m f i r i n g a s l o n g a st h e f e a t u r e s a r e c a p a b l e o f u n i f ic a t i o n .O n e p o s s i b i l i t y w o u l d b e t o a l l ow t h e ( e r r o r+ ) f e a t u r e t o p e r c o l a t e u p t h r o u g h t h e p a r s e .A n y ru l e w h i c h i n t r o d u c e s t h e ( e r r o r + ) fe a -t u r e c o u l d t h e n b e p r e v e n t e d f r o m h a v i n g a n yc h i l d re n s p e c i fi e d w i t h ( e r r o r + ) . H o w e v e r ,t h i s s o l u t i o n w o u l d b e f a r t o o r e s t r i c t i v e , a s i tw o u l d r e s t r i c t t h e n u m b e r o f e r r o rs i n a s e n t e n c et o o n e , a n d m a n y o f t h e s e n t e n c e s i n o u r A S Lc o r p u s i n v o l v e m u l t i p l e e r r o r s .R e c a l l, h o w e v e r , t h a t i n o u r a n a l y s i s w e f o u n dt h a t ( e x c e p t fo r d e t e r m i n e r s ) o u r w r i t i n g s a m -p l e s d i d n o t c o n t a i n m u l t i p l e o m i s s i o n e r ro r s i na s e n t e n ce . T h u s a n o t h e r p o s s i b i l i t y m i g h t b e t op e r c o l a t e a n e r r o r f e a t u r e a s s o c i a t e d w i t h o m i s -s io n s o n l y - p e r h a p s c al le d ( m i s s i n g + ) .

U p o n c l o se r i n s p e c t io n , t h i s s o l u t i o n a l s o h a sd i f fi c u l ti e s . T h e f i r s t d i ff i c u l t y h a s t o d o w i t hi m p l e m e n t i n g t h e f e a t u r e p e r c o l a t i o n . F o r i n-s t a n c e , f o r a V P t o b e s p e ci f ie d as ( m i s s i n g+ ) w h e n e v e r a n y o f i ts s u b - c o n s t i t u e n t s h a s t h a tf e a t u r e, o n e w o u l d n e e d t o h a v e s e p a r a t e r u le sr a i si n g t h e f e a t u r e u p f r o m e a c h o f t h e s u b -c o n s t i t u e n t s , a s i n t h e f o l l o w i n g :( 8 ) V P ( m i s s i n g ? a ) ~ V N P N P ( m i s s i n g ? a )(9) VP(mi ssin g ?a) --~ V NP(m iss ing ?a) NP(I0) V P(m issi ng ?a) --> V(m iss ing ?a) NP NPT h i s w o u l d c a us e a n u n w a r r a n t e d i n cr e as e i nt h e s iz e o f t h e g r a m m a r , a n d w o u l d a l so c a us e

a n i m m e n s e i n cr e as e i n t h e n u m b e r o f p ar se s,s i nc e t h r e e V P s w o u l d b e a d d e d t o t h e c ha rt ,one for each of the rules.A t f ir s t g l a n c e it a p p e a r s t h a t t h i s p r o b l e mc a n b e o v e r c o m e w i t h t h e u s e o f " f o o t fe a t u r e s ,"w h i c h a r e i n c l u d e d i n t h e p a r s e r w e a r e u s in g . Af o o t f e a t u r e m o v e s f e a t u r e s f r o m a n y c h i l d t o t h ep a r e n t . F o r e x a m p l e , f o r a f o o t fe a t u r e F , i f o n ec h i l d h a s a s p e c i f i c a t i o n f o r F , i t w i l l b e p a s s e d

1 2 0 0

7/27/2019 P98-2196

4/7

o n t o t h e p a r e n t . I f m o r e t h a n o n e c h i ld i s s p e c -i fi e d f o r F , t h e n t h e v a l u e s o f F m u s t u n i f y , a n dt h e u n i f i e d v a l u e w i l l b e p a s s e d u p t h e p a r e n t .W h i l e t h e u s e o f f o o t f e a t u re s a p p e a r s t o m a k et h e f e a t u r e p e r c o l a t i o n e a s i e r , i t w i l l n o t a l l o wt h e f e a t u r e t o b e u s e d a s d e s i re d . I n p a r t i c u -l ar , w e n e e d t o h a v e t h e f e a t u r e p e r c o l a t e d o n l yw h e n i t h a s a p o s i t i v e v a l u e a n d o n l y w h e n t h a tv a l u e i s a s s o c i a t e d w i t h e x a c t l y o n e c o n s t i t u e n to n t h e r i g h t - h a n d s i d e o f a ru l e . T h e f o o t f e a-t u r e a s d e f i n e d b y t h e p a r s e r w o u l d a l l o w t h ep e r c o l a t i o n o f t h e f e a t u r e e v e n i f i t w e r e s p e c i-f i e d i n m o r e t h a n o n e c o n s t i t u e n t .A f u r t h e r c o m p l i c a t i o n w i t h u s i n g t h i s t y p eo f f e a t u r e p r o p a g a t i o n a r is e s b e c a u s e t h e r e a r es o m e s i t u a t i o n s w h e r e m u l t i p l e o m i s s i o n e r r o r sd o o c c u r , e s p e c i a ll y w h e n d e t e r m i n e r s a r e o m i t -t e d . 1 C o n s i d e r t h e f o ll o w in g e x a m p l e t a k e nf ro m o u r c o r p u s w h e r e b o t h t h e m a i n v e r b " b e "a n d a d e t e r m i n e r " t h e " a r e o m i t t e d .( 1 1 ) S t u d e n t a l w a y s b o t h e ri n g m e w h i le I a ma t d o r m .( C o r r e c t e d ) S t u d e n t s a r e a l w a y s b o t h e r i n g m ew h il e I a m a t th___.eed o r m .O u r s o l u t i o n t o t h e p r o b l e m i n v o l v e s u s i n gp r o c e d u r a l a t t a c h m e n t . T h e p a r s e r w e a r e u s -i n g b u i l d s c o n s t i t u e n t s a n d s t o r e s t h e m i n ac h a r t. B e f o r e s t o r i n g t h e m i n t h e c h a rt , t h ep a r s e r c a n r u n a r b i t r a r y p r o c e d u r e s o n n e w c o n -s t i t u e n t s . T h e s e p r o c e d u r e s , s p e c i fi e d i n t h eg r a m m a r , w i ll b e r u n o n a l l c o n s t i t u e n t s t h a tm e e t a c e r t a i n p a t t e r n s p e c if i ed b y th e g r a m -m a r w r i t e r .

O u r p r o c e d u r e a m o u n t s t o s p e c i fy i n g a n a l-t e r n a ti v e m e t h o d f o r p r o p a g a t i n g t h e ( m i s s i n g+ ) f e a t u r e , w h i c h w i l l s t i ll b e a f o o t f e a t u r e .I t w i ll b e r u n o n a n y c o n s t i t u e n t t h a t s p e c if ie s( m i s s i n g + ) . T h e p r o c e d u r e c a n e i th e r d el e tea c o n s t i t u e n t t h a t h a s m o r e t h a n o n e c h i ld w i t h( m i s s i n g + ) , o r i t c a n a l te r t h e ( m i s s i n g + )f e a t u r e o n t h e c o n s t i t u e n t i n t h e f a c e o f d e t e r -m i n e r o m i s s io n s ( a s d is c u s s e d i n f o o t n o t e 1 ). B yu s i n g a s p e c i a l p r o c e d u r e t o i m p l e m e n t t h e f e a-t u r e p e r c o l a t i o n , w e w il l b e a b l e t o b e m o r e f le x -i b l e i n w h e r e w e a l lo w t h e " m i s s i n g " f e a t u r e t op e r c o l a t e .3 .3 S y n t a c t i c F e a t u r e A d d i t i o nF o r t h i s s y s t e m t o p r o p e r l y m o d e l l a n g u a g e a c -q u i s it i o n , i t m u s t a l s o m o d e l t h e a d d i t i o n ( a n dp o s s i b l e s u b t r a c t i o n ) o f s y n t a c t i c f e a t u r e s i n t h el e xi c o n a n d g r a m m a r o f t h e l e a rn e r . F o r i n-s t a n c e , A S L n a t i v e s h a v e a g r e a t d e a l o f d i f-f i c u lt y w i t h m a n y o f t h e a g r e e m e n t f e a t u r e s i n

1While our analysis so far has only indicated thatdeter miner omissions have this property, we do not w antto rule out the possibility that other combinations ofomission errors might be found to occur as well.

E n g l is h . A s a c o n c r e t e e x a m p l e , t h i s p o p u l a t i o nf r e q u e n t l y h a s t r o u b l e w i t h t h e d i f f e r e n c e b e -t w e e n "o t h e r " a n d " a n o t h e r " . T h e y f r e q u e n t l yu s e " o t h e r " i n a s i n g u l a r N P , w h e r e " a n o t h e r "w o u l d n o r m a l l y b e c a l le d fo r. W e h y p o t h e s i z et h a t t h i s i s p a r t l y a r e s u l t o f t h e i r n o t u n d e r -s t a n d i n g t h a t t h e r e is a g r e e m e n t b e t w e e n N P sa n d t h e i r s p e c i f i e rs (d e t e r m i n e r s , q u a n t i f i e r s ,e t c . ). E v e n i f t h i s is r e c o g n i z e d , t h e l e a r n e r sm a y n o t h a v e t h e l e xi c a l r e p r e s e n t a t i o n s n e c -e s s a r y to s u p p o r t t h e a g r e e m e n t f o r t h e s e t w ow o r d s . 2 T h u s , t h e m o s t a c c u r a t e m o d e l o f t h el a n g u a g e o f t h e s e e a r l y l e a r n e r s i n v o l v e s a l e xi -c o n w i t h i m p o v e r i s h e d e n t r ie s - i .e . n o p e r s o no r n u m b e r f e a t u r e s fo r d e t e r m i n e r s a n d q u a n t i -t ie rs . S u c h a n i m p o v e r i s h e d l e x i c o n w o u l d m e a nt h a t t h e e n t r ie s f o r t h e t w o w o r d s m i g h t b e i d e n -t i c a l , w h i c h a p p e a r s t o b e t h e c a s e f o r t h e s el e a r n e r s .

T h e r e a r e a t l e a s t t w o r e a s o n s f o r n o t u s -i n g t h i s s o r t o f i m p o v e r i s h e d l e x i co n . F i r s t l y ,i t w o u l d r e q u i r e h a v i n g m u l t i p l e l e x i c o n s (s o m ei m p o v e r i s h e d , o t h e r s n o t ) , w i t h t h e s y s t e mn e e d i n g t o d e t e r m i n e w h i c h t o u s e f o r a g i v e nu s e r . S e c o n d l y , i t w o u l d n o t a l lo w g r a m m a t -i c a l u s e s o f t h e i m p o v e r i s h e d i t e m s t o b e d i f -f e r e n t i a te d f ro m u n g r a m m a t i c a l u s e s . W i t h a ni m p o v e r i s h e d l e x i c o n , a n y u s e ( g r a m m a t i c a l o rn o t ) o f " o t h e r" o r " a n o t h e r " w o u l d b e f la g g e da s a n e r r o r , s i n c e i t w o u l d i n v o l v e u s i n g a l e x ic a le n t r y t h a t d o e s n o t h a v e a l l o f t h e f e a t u r e s t h a tt h e s t a n d a r d e n t r y h a s . S i n c e t h e l e x ic a l i t e mw o u l d n o t h a v e t h e a g r s p e c i f i c a t i o n , i t c o u l dn o t m a t c h t h e r u le t h a t r e q u i r e s a g re e m e n t b e -t w e e n d e t e r m i n e r s a n d n o u n s .3 . 3 . 1 I m p l e m e n t a t i o nF o r t h e s e r e a s o n s , w e d e c i d e d n o t t o u s e d i f f e r -e n t l e x i c a l e n t r i e s t o m o d e l t h e d i f f e r e n t s t a g e so f a c q u i s i t i o n . I n s t e a d , w e u s e m a l - r u l e s , t h es a m e m e c h a n i s m t h a t w e a r e u s i ng t o m o d e ls y n t a c t i c c h a n g e s. A s t a n d a r d ( g r a m m a t i c a l )D P ( D e t e r m i n e r P h r a s e ) r u le h a s th e f o ll o w i n gf o r m a t :

(12) DP(agr ?a) --~ Det ( agr ?a) NP(agr ?a)W e in i t i a l ly t r i e d s i m p l y e l i m i n a t i n g t h e r e f -

e r en c e s t o a g re e m e n t b e t w e e n t h e N P a n d t h ed e t e r m i n e r , a s i n t h e f o l l o w i n g m a l - ru l e :(13) DP( erro r +)( agr ?a) --+ Det NP(agr ?a)

T h i s h a s t h e a d v a n t a g e o f f la g g in g a n y d e -v i a n t D P s a s h a v i n g t h e e r r o r f e a t u re , s i n c e u n -g r a m m a t i c a l D P s w i l l t r i g g e r t h e m a l - r u l e ( 1 3 ),b u t w o n ' t t ri g g e r ( 12 ). H o w e v e r , a g r a m m a t i c a l2 "Another" and "other" are n ot separate lexical itemsin ASL.

1 2 0 1

7/27/2019 P98-2196

5/7

D P ( e .g . " a n o t h e r c h i l d " ) fi r es b o t h t h e m a l -r u l e (1 3 ) a n d t h e g r a m m a t i c a l r u l e ( 1 2 ). N o to n l y d i d t h i s b e h a v i o r c a u s e t h e p a r s e r t o s l o wd o w n v e r y s i g n i f ic a n t l y , s i n c e i t e f f e c t i v e ly d o u -b l e d t h e n u m b e r o f D P s i n a s e n t e n c e , b u t i t a l soh a s t h e p o t e n t i a l t o r e p o r t a n e r r o r w h e n o n ed o e s n o t e x i s t . W e a ls o b r ie f l y c o n s i d e r e d u s i n gi m p o v e r i s h m e n t r u l e s o n s p e c if i c c a t e g o r i e s . F o re x a m p l e , w e c o u l d h a v e u s e d a r u le s t a t i n g t h a td e t e r m i n e r s h a v e a l l p o s s i b l e a g re e m e n t v a lu e s .T h i s h a s t h e e f fe c t o f e l i m i n a t i n g a g r e e m e n t a sa b a r r i e r t o u n i f i ca t i o n , m u c h a s w o u l d b e e x -p e c t e d i f t h e l e a r n e r h a s n o k n o w l e d g e o f a g r ee -m e n t o n d e t e r m i n e r s . H o w e v e r , t h i s s o l u t i o nh a s a p r o b l e m v e r y s im i l a r t o t h a t o f t h e p r e -v i o u s p o s s i b l e s o l u t i o n : a l l d e t e r m i n e r s i n t h ei n p u t c o u l d s u d d e n l y h a v e t w o e n t r i e s i n t h ec h a r t - o n e w i t h t h e a c t u a l a g r e e m e n t , o n e w i t ht h e i m p o v e r i s h e d a g r e e m e n t . T h e s e w o u l d t h e nb o t h b e u s e d i n p a r s i n g , l e a d i n g t o a n o t h e r e x -p l o s i o n in t h e n u m b e r o f p a r s e s.W e f i n a ll y e n d e d u p b u i l d i n g a s e t o f r u le st h a t m a t c h e s j u s t t h e u n g r a m m a t i c a l p o s s i b i li -t i e s, i. e. t h e y d o n o t a l l o w a g r a m m a t i c a l s t r u c -t u r e t o f ir e b o t h t h e m a l - r u l e a n d t h e n o r m a lr u le . T h e p r e s e n t s e t o f r u l e s f o r d e t e r m i n e r -N P a g r e e m e n t i n c l u d e t h e f o l l o w i n g :(14) DP(agr ?a) --+ Det (agr ?a) NP (agr? a )( 1 5 ) D P ( a g r s ) ( e r r o r + ) -+ D e t ( a g r ( ? ! as ) ) N P ( a g r s )( 1 6 ) D P (a g r p ) ( e r r o r + ) ~ D e t ( a g r ( ? ! ap ) ) N P ( a g r p )

T h i s s o l u t i o n r e q u i r e d u s i n g t h e n e g a t i o n o p -e r a t o r " !" p r e s e n t in o u r p a r s e r t o s p e c i f yt h a t a D e t n o t a l l ow s in g u l a r / p l u r a l a g r e e m e n t .H o w e v e r , t h i s f e a t u r e i s l i m i t e d i n t h e p r e s e n ti m p l e m e n t a t i o n t o c o n s t a n t v a lu e s , i .e . w ec a n ' t n e g a t e a v a r i a b l e . T h i s s o l u t i o n a c h i e v e st h e m a j o r g o a l o f n o t i n t r o d u c i n g e x t r a n e o u sp a r s e s f o r g r a m m a t i c a l c o n s t i t u e n t s . H o w e v e r ,i t a c h ie v e s t h i s g o a l a t s o m e c o s t. N a m e l y , w ea r e f o r c e d t o i n c r e a se t h e n u m b e r o f r u le s i n o r-d e r t o a c c o m p l i s h t h e t a s k .3 . 3 . 2 F u t u r e p l a n sW e a r e p r e s e n t l y w o r k i n g o n t h e i m p l e m e n t a -t i o n o f a v a r i a n t o f u n i f i c a t i o n t h a t w i l l a l lo w u st o d o t h e jo b w i t h f e w e r r u le s . T h e n e w o p e r a -t i o n w i l l w o r k i n t h e f o l lo w i n g s o r t o f r u le :

( 1 7 ) DP (agr ?a)--+ Det (agr ?!a) NP(agr ?a)T h i s r u l e w i ll b e i n t e r p r e t e d a s fo l lo w s : t h ea g r v a l u e s b e t w e e n t h e D P a n d t h e N P w i ll b et h e s a m e , a n d n o n e o f t h e v a l u e s in D e t w i l lb e a l l o w e d t o b e i n t h e a g r e e m e n t v a l u e s f o r

t h e N P a n d t h e D P. T h i s w i l l a l l o w t h e r u l e t of ir e p r e c i se l y w h e n t h e r e a r e n o p o s s i b l e w a y st o u n i f y t h e v a l u e s b e t w e e n t h e D e t a n d t h e N P,i .e . n o n e o f t h e a g r v a l u e s f o r t h e D e t w i l l b ea l l o w e d in t h e v a r i a b l e ? a . T h u s , t h i s r u l e w i llo n l y fi re f o r u n g r a m m a t i c a l c o n s t r u c t i o n s .4 G r a m m a r C o v e r a g e / U s e r I n t e r f a c eT h e I C I C L E g r a m m a r is a b ro a d - c o v e ra g eg r a m m a r d e s i g n e d t o p a r s e a w i d e v a r ie t y o fb o t h g r a m m a t i c a l s e n t e n c e s a n d s e n t e n c e s c o n -t a i n i n g er r o rs . I t i s b u i l t a r o u n d t h e C O M -L E X S y n t a x 2 . 2 l e x i c o n ( G r i s h m a n e t a l . , 1 9 9 4 ) ,w h i c h c o n t a i n s a p p r o x i m a t e l y 3 8 , 0 0 0 d i f f e r e n ts y n t a c t i c h e a d w o r d s . W e h a v e a s i m p l e s e to f r u l e s t h a t a l l o w s f o r i n f l e c ti o n , t h e r e b y d o u -b l i n g t h e n u m b e r o f n o u n f o r m s , w h i l e g iv i n g u st h r e e t o f o u r t i m e s a s m a n y v e r b f o r m s a s t h e r ea r e h e a d s . T h u s w e c a n h a n d l e a p p r o x i m a t e l y4 0 , 0 0 0 n o u n f o r m s , 8 , 0 0 0 a d j e c t i v e s , a n d w e l lo v e r 1 5 ,0 0 0 v e r b f or m s . I n a d d i t i o n , u n k n o w nw o r d s c o m i n g i nt o t h e s y s t e m a r e a s s u m e d t ob e p r o p e r n o u n s, t h u s e x p a n d i n g t h e n u m b e r o fw o r d s h a n d l e d e v e n f u r t h e r .T h e g r a m m a r i t se l f c o n t a in s a p p r o x i m a t e l y2 5 d i f f e r e n t a d j e c t i v a l s u b c a t e g o r i z a t i o n s , i n -c l u d i n g s u b c a t e g o r i z a t i o n s r e q u i r i n g a n e x t r a -p o s e d s t r u c t u r e ( t h e " i t " in " i t i s t r u e t h a th e i s h e r e " ) . W e a l s o i n c l u d e h a l f a d o z e nn o u n c o m p l e m e n t a t i o n ty p e s . W e h a v e a p -p r o x i m a t e l y 1 1 0 d i f fe r e n t v e r b c o m p l e m e n t a t i o nf r am e s , m a n y o f w h i c h a r e i n d e x e d f o r s e v e ra ld i ff e re n t s u b c a t e g o r i z a t i o n s . T h e g r a m m a r isa l s o a b l e t o a c c o u n t f o r v e r b - p a r t i c l e c o n s t r u c -t i o n s w h e n t h e v e r b i s a d j a c e n t t o t h e p a r t i c l e ,a s w e ll a s w h e n t h e y a r e s e p a r a t e d ( e .g . " I c a l l e dh i m u p " ) .A d d i t i o n a l l y , t h e g r a m m a r a l l o w s f o r v a r i o u sd i f f e r e n t t y p e s o f s u b j e c t s , i n c l u d i n g i n f i n i ti v a l sw i t h a n d w i t h o u t s u b j e c t s ( " t o f a il a c la s s i su n f o r t u n a t e " , " f o r h i m t o f a il t h e c l a s s is i r re -s p o n s i b l e " ) . I t h a n d l e s y e s / n o q u e s t i o n s , w h -q u e s t i o n s , a n d b o t h s u b j e c t a n d o b j e c t r e l a t i v ec l a u s e s .T h e g r a m m a r h a s o n l y l i m i t e d a b i li t i es c o n -c e r n in g c o o r d i n a t i o n - i t o n l y a ll o w s l i m i t e dc o n s t i t u e n t c o o r d i n a t i o n , a n d d o e s n o t a ll o wn o n - c o n s t i t u e n t c o o r d i n a t i o n ( e .g . " I s a w a n dh e h i t t h e b a l l " ) a t a l l. I t i s a l s o fa i r l y w e a ki n i ts h a n d l i n g o f a d j u n c t s u b o r d i n a t e c l a u s e s.T h e p o p u l a t i o n w e a r e c o n c e r n e d w i t h a l s o h a ss i g n i f ic a n t t r o u b l e w i t h t h i s , i n p a r t i c u l a r t h e r ei s a s t r o n g p r o p e n s i t y t o w a r d s o v e r - u s i n g " b e -c a u s e " . A d v e r b s a r e a l so p r o b l e m a t i c , i n t h a tt h e s y s t e m i s n o t y e t a b l e t o d i ff e r e n t ia t e w h a tp o s i t i o n a g i v en a d v e r b s h o u l d b e a b l e t o ta k e i na s e n t en c e , t h u s n o e r ro r s i n a d v e r b p l a c e m e n t

1 2 0 2

7/27/2019 P98-2196

6/7

can be flagged. We are presently in the processof integrating a new version of the lexicon thatincludes features specifying what each adverbcan atta ch to. Once this is done, we expect tobe able to process adverbs quite effectively.The user interface presently consists of a mai nwindow where the user can input the text andcontrol parsing, file access, etc. After parsing,the sentences are highlighted with different col-ors corresponding to different types of errors.When the user double-clicks on a sentence, aseparate "fix-it" window is displayed with thesentence in question, along with descriptions ofthe errors. The user can click on the errors andthe system will highlight the part of the sen-tence where the error occurred. For example,in the sentence "I see a boys", only "a boys"will be highlighted. The "fix-it" window alsoallows the user to change the sentence and thenre-parse it. If the changes are acceptable to theuser, the new sentence can be substituted backinto the main text.5 Ev a lu a t i o n o f E r ro r Re c o g n i t io nAn evaluation of the grammar was conductedon a variety of sentences pulled from the cor-pus of ASL natives. The corpus co ntains essayswritten by ASL natives which is annot ated wi threferences to different types of errors in the sen-tences. The focus for this paper was on recog-nition of agreement-type problems, and as suchwe pulled out all of the sentences that had beenmarked with the following errors: NUM: Number problems, which are typi-cally errors in subject-verb agreement

ED: extra determiner MD: missing deter mine r for an NP th at re-quires a determiner ID: incorrect determinerIn addition to testing sentences with theseproblems, we also tested fully grammatical sen-tences from th e same corpus, to see if we couldcorrectly differentiate between gramm atical andungrammatical sentences that might be pro-

duced by our target user group.After gathering the sentences from thedatabase, we cut them down to mono-clausalsentences wherever possible, due to the fact th atthe handling of adjunct clauses is not yet com-plete (see 4). An example of the typ e of sen-tence that had to be divided is the following:(18) T h e y s h o u l d c o m m u n i c a t e e a c h o t h e r b e -c a u se t h e c o m m u n i c a t i o n i s v e r y i m p o r t a n t t ou n d e r s t a n d e a c h o t h e r .

This sentence was divided into "They shouldcommunicate each other" and "the communi-cation is very important to understand eachother." In addition to separating the clauses,we also fixed the spelling errors in the sentencesto be tested since spelling correction is beyondthe scope of the current implementation.5 .1 Re s u l t s fo r Un g ra mma t ic a l

S e n t e n c e sWe ended up with 79 sentences to test for thedeter miner and agreement errors. Of these 79sentences, 44 (56%) parse with the expectedtype of error. Anot her 23 (29%) have no parsesthat cover the entire sentence, and 12 (15%)parse as having no errors at all.A number of the sentences tha t had beenflagged with errors in the data base were actual lygrammatical sentences, but were deemed inap-propri ate in context. Thus, sentences like thefollowing were tagged wit h errors in the corpus:

(19) I s t a r t e d t o a t t e n d t h e c l a s s l a s t S a t u r d a y .It was evident from the context that this sen-tence should have had "classes" rather than"the class." Of the 12 sentences th at wereparsed as error-free, five were actually syntacti-cally and semantically acceptable, but were in-appropriate for their contexts, as in the previousexample. Another four had pragmatic/semanticproblems, but were syntac tica lly well-formed, asin

(20) I w a n t t o s u c c e e d in jo b s a n y w h e r e .Thus, there are really only three sentencesthat do not have a parse with the appropriateerror. Since this parser is a syntacti c parser,it should not be expected to find the seman-tic/ pragm atic errors, nor should it know if thesentence was inappropriate for its context in theessay. If we elimin ate the nine sentences th atare actually grammatical in isolation, we areleft with 70 sentences, of which 44 (63%) haveparses with the expected error, three (4%) arewrongly accepted as grammati cal, and 23 (33%)do not parse.In terms of evaluating these results for the

purposes of the system, we must consider theimplicati ons of the various categories. 63%would trigger tutoring, and 33% would betagged as problematic, but would have no in-formation about the type of error. In only 4%of sentences containing errors would the systemincorrectly indicate that no errors are present.5 .2 Resu l ts fo r G r a m m a t i c a l S e n t e n c e sWe also tested the system on 101 grammaticalsentences that were pulled from the same cor-pus. These sentences were modified in the same

1203

7/27/2019 P98-2196

7/7

way as the ungrammatical ones, with multi-clausal sentences being divided up into mono-clausal sentences. Of these 101 sentences , 89(88%) parsed as having no errors, 3 (3%) parsedwith errors, and the remaining 8 (8%) did notparse.The present implementation of the grammarsuffers from poor recognition of coordination,even within single clauses. Five of the elevensentences that did not r eturn an error-free parsesuffered from this limitation. We expect to beable to improve the numbers significantly byincluding in the grammar some recognition ofpunctuation, which, due to technical problems,is presently filtered out of the input before theparser has a chance to use it.6 C onc l u s i on s a nd F u t u r e W or kFuture work will include extending the gram-mar to better deal with coordination and ad-junct clauses. We will also continue to work onthe negation operator and the propagation ofthe mis si ng feature discussed above. In orderto cut down on the number of parses, as well asto make it easier to decide which is the appropri -ate parse to correct, we have recently switchedto a best-first parsing strategy. This should al-low us to model which rules are most likely tobe used by a given user, with the mal-rules cor-responding to the constructions currently beingacquired having a higher probability than thosethat the learner has already mastered. How-ever, at the moment we have simply lowered theprobabilities of all mal-rules, so that any gram-matical parses are generated first, followed bythe "ungrammatical" parses.As we have shown, this system does a goodjob of flagging ungrammatical sentences pro-duced by the target population, with a highproportion of the flagged sentences containingsignificant information about the type and lo-cation of the error. Our continuing work willhopefully improve these percentages, and couplethis recognition component with an intelligenttutoring phase.R e f e r e nc e sJames Allen. 1 99 5 . Natural LanguageUnderstanding, Second Edition. Ben-jamin/Cummings, CA.N. Bailey, C. Madden, and S. D. Krashen . 1974.Is there a 'natural sequence' in adult sec-ond language learning? Language Learning,24(2):235-243.Vivian Cook. 1993 . Linguis t ics and SecondLanguage Acquisi t ion. Macmillan Press Ltd,London.

Heidi C. Dulay and Marina K. Burt. 1974. Nat-ural sequences in child second language acqui-sition. Language Learning, 24:37-53.Rod Ellis. 1994 . The Stud y o f Second Lan-guage Acquisition. Oxford University Press,Oxford.Ralph Grishman, Catherine Macleod, andAdam Meyers. 1994. Comlex syntax: Build-ing a computational lexicon. In Proceedingso f the 15th In ternat iona l Conference on Com-putational Linguistics, Kyoto, Japan, July.Coling94.C.-T. James Huang. 1984. On the distributionand reference of empty pronouns. Linguis t icInquiry, 15(4):531-574, Fall.David Ingram. 1989. First Language Acqui -si t ion: Method, Description, and Explana-tion. Cambridge University Press, Cam-bridge; New York.Stephen Krashen. 198 1. Second LanguageAcquisi t ion and Second Language Learning.Pergamon Press, Oxford.Diane C. Lillo-Martin. 1991. Universal Gram-ma r a n d Amer i c a n S i gn L a n gu a ge . KluwerAcademic Publishers, Boston.Lisa N. Michaud and Ka thle en F. McCoy. 1998.Planning tutorial text in a system for teach-ing english as a second language to deaf learn-ers. In Pro ceedi ngs o f t he 1 9 9 8 A A A I W o rk -shop on Integrating Arti f icial Intel l igence andAssi s t ive Technology , Madison, Wisconsin,July.

Lisa N. Michaud. 1998. Tutorial response gen-eration in a writing tool for deaf learnersof english. In Proceed ings o f the Fi f teen thNational Conference on Arti f icial Intel l igence(poster abstract), Madison, Wisconsin, July.L. Selinker. 1972. Interlanguage. I n t e rn a t i o n a lRev iew o f Appl ied L inguis t ics , 10:209-231.D. Sleeman. 1982. Inferring (mal) rules frompupil's protocols. In Proceedings o f ECAI-82 ,pages 160-164, Orsay, France. ECAI-82.Linda Z. Suri and Kat hleen F. McCoy. 1993. Amethodology for developing an error taxon-omy for a computer assisted language learn-ing tool for second language learners. Techni-cal report TR-93-16. Dept. of CIS, Universityof Delaware.Lev Semenovich Vygotsky. 1986. Thought andLanguage. MIT Press, Cambridge, MA.Ralph M. Weischedel, Wilfried M. Voge, andMark James. 1978. An artifi cial intelligenceapproach to language instruction. Art i f i c ia lIntelligence, 10:225-240.

1204