In case analysis, two ways are generally used for parsing.. In this expression, the structure can be described by using case frames... Therfore, the predicted phrase is tested to determi
Trang 1A Case A n a l y s i s Method C o o p e r a t i n g with ATNG
and Its A p p l i c a t i o n to Machine Translation
H i t o s h i IIDA, K e n t a r o O G U R A and Hirosato N O M U R A
Musashino Electrical Communication Laboratory, N.T.T
Musashino-shi, Tokyo, 180, Japan
Abstract
This paper present a n e w m e t h o d for parsing
English sentences T h e parser called L U T E - E J parser
is combined with case analysis a n d A T N G - b a s e d
analysis L U T E - E J parser has two interesting
mechanical characteristics O n e is providing a
structured buffer, Structured Constituent Buffer, so
as to hold previous fillers for a case structure, instead
of case registers before a verb appears in a sentence
T h e other is extended H O L D mechanism(in ATN), in
whose use an e m b e d d e d clause, especially a "be-
deleted" clause, is recursively analyzed by case
analysis This parser's features are (1)extracting a
case filler, basically as a n o u n phrase, by A T N G -
based analysis, including recursive case analysis, and
(2)mixing syntactic and semantic analysis by using
case frames in case analysis
I Introduction
In a lot of natural language processing including
machine translation, A T N G - b a s e d analysis is a usual
method, while case analysis is c o m m o n l y employed
for J a p a n e s e l a n g u a g e processing.The parser
described in this paper consists of two major parts
O n e is A T N G - b a s e d analysis for getting case
elements and the other is case-analysis for getting a
semantic clause analysis L U T E - E J parser has been
i m p l e m e n t e d o n a n e x p e r i m e n t a l m a c h i n e
translation system L U T E (Language Understander,
Translator & Editor) which can translate English
into Japanese a n d vice versa L U T E - E J is the
English-to-Japanece version of L U T E
In case analysis, two ways are generally used for
parsing O n e w a y analyzes a sentence from left to
right, by using case registers Case fillers which fill
each case registers are m a j o r participants of
constituents, for e x a m p l e S U B J E C T , O B J E C T ,
PP(Prepositional Phrase)'s and so on, in a sentence
In particular, before a verb appears, at least one
participant(the subject) will be registered, for
example, in the A G E N T register
T h e o t h e r m e t h o d h a s two p h a s e s on t h e a n a l y s i s
p r o c e s s i n g I n t h e f i r s t p r o c e s s i n g , p h r a s e s a r e
e x t r a c t e d as case e l e m e n t s in o r d e r to fill t h e slots of a case f r a m e T h e second is to choose t h e a d e q u a t e case
e l e m e n t a m o n g t h e e x t r a c t e d p h r a s e s for a c e r t a i n case slot a n d to c o n t i n u e t h i s process for t h e o t h e r
p h r a s e s a n d t h e o t h e r c a s e slots I n t h i s m e t h o d ,
t h e r e a r e no special actions, i.e no r e g i s t e r i n g before
a v e r b a p p e a r s ( W i n o g r a d [83] )
E n g l i s h q u e s t i o n - a n s w e r i n g s y s t e m P L A N E S ( W a l t z [78] ) u s e s a s p e c i a l k i n d of c a s e f r a m e s ,
" c o n c e p t case f r a m e s " By u s i n g t h e m , p h r a s e s i n a
s e n t e n c e , w h i c h a r e d e s c r i b e d by u s i n g p a r t i c u l a r
" s u b n e t s " and s e m a n t i c f e a t u r e s (for a p l a n e type a n d
so on), a r e g a t h e r e d a n d a n action of a r e q u i r e m e n t (a
s e n t e n c e ) is constructed
2 L U T E - E J P a r s e r 2.1 L U T E - E J P a r s e r ' s D o m a i n
T h e d o m a i n t r e a t e d by L U T E - E J p a r s e r is w h a t
m i g h t be c a l l e d a s e t o f " c o m p l e x s e n t e n c e s a n d
c o m p o u n d s e n t e n c e s " L e t S be a n e l e m e n t of t h i s s e t
a n d l e t C L A U S E be a s i m p l e s e n t e n c e (which m i g h t
i n c l u d e a n e m b e d d e d sentence) N o w , i f M A J O R - C L
a n d M I N O R - C L a r e p r i n c i p a l c l a u s e a n d s u b o r d i n a t e clause, r e s p e c t i v e l y , S c a n be w r i t t e n as follows (R1} <S > :: = (< MINOR-CL >) < MAJOR-CL >
(<MINOR-CL>) (R2) <MAJOR-CL>::= <CLAUSE> / <S> (R3) <MINOR-CL>::= <CONJUNCTION>
<CLAUSE> (in BNF) The syntactic and semantic structure for a CLAUSE is basically expressed by a case structure
In this expression, the structure can be described by using case frames The described structure implies
t h e s e m a n t i c s t r u c t u r e i n t e n d e d by a C L A U S E a n d
m a i n l y d e p e n d i n g on v e r b l e x i c a l i n f o r m a t i o n
C a s e e l e m e n t s in a C L A U S E are N o u n P h r a s e s , object N P s of P P s or s o m e k i n d s of A D V e r b s w i t h
r e l a t i o n to t i m e s a n d locations T h e N P s t r u c t u r e is described as follows,
(R4) < N P > :: = (<NHD >){ < NP>/NOUN}( < NMP >) / < Gerund-PH > / < To-infmitive~PH > /That < CLAUSE >
Trang 2w h e r e N H D ( N o u n H e a D e r ) is ~premodification" a n d
N M P ( N o u n Modifier P h r a s e ) is " p o s t m o d i f i c a t i o n '
T h u s , N M P is a s e t i n c l u d i n g v a r i o u s k i n d s o f
e m b e d d e d f i n i t e c l a u s e s , r e l a t i v e or b e - d e l e t e d
r e l a t i v e f i n i t e clauses
A f t e r m o r p h o l o g i c a l a n a l y s i s w i t h l o o k i n g u p
words for a n i n p u t s e n t e n c e in t h e d i c t i o n a r y , a n
i n p u t sentence a n a l y s i s is b e g u n from left to r i g h t
Thus, a f t e r a verb h a s been seen, i t m a k e s progress to
a n a l y z e a C L A U S E by r e f e r r i n g to t h e case f r a m e
c o r r e s p o n d i n g to the verb, a s each slot in t h e case
f r a m e is filled w i t h a n N P or an object of PP A case
slot consists of t h r e e elements: one s e m a n t i c f i l l e r
condition slot a n d two s y n t a c t i c a n d s e m a n t i c m a r k e r
s l o t s H e r e , a p r e p o s i t i o n is d i r e c t l y u s e d a s a
s y n t a c t i c m a r k e r F u r t h e r m o r e , f o u r p s e u d o
m a r k e r s , ~subject", "object", ~ i n d i r e c t - o b j e c t " a n d
~complement", a r e used A s a s e m a n t i c m a r k e r , a so-
called deep case is used (now, 41 r e a d y for t h i s case
s y s t e m ) T h e n , L U T E - E J P a r s e r e x t r a c t s t h e
s e m a n t i c s t r u c t u r e i m p l i e d in a s e n t e n c e (S o r
C L A U S E ) a s an e v e n t or s t a t e i n s t a n c e c r e a t e d from
a case frame, which is a class or a prototype A n N P is
p a r s e d by the A T N G - b a s e d a n a l y s i s in order to decide
a case slot filler {now, 81 nodes on t h i s ATNG)
N e x t , t h e r e a s o n w h y t h e c a s e a n a l y s i s a n d
A T N G - b a s e d a n a l y s i s are m e r g e d will be s t a t e d I t
h a s two m a i n points
O n e p o i n t i s a b o u t t h e d e p t h o f e m b e d d e d
s t r u c t u r e s F o r e x a m p l e , t h e i n v e s t i g a t i o n on t h e
d e g r e e of a C L A U S E c o m p l e x i t y r e s u l t e d in t h e
necessity to h a n d l e a h i g h degree of c o m p l e x i t y w i t h
efficiency The N M P s t r u c t u r e is also m o r e complex
I n p a r t i c u l a r , e m b e d d e d V P s or A D J P H s a p p e a r
r e c u r s i v e l y T h e r e f o r e , a r e c u r s i v e p r o c e s s f o r
a n a l y z i n g N P is needed
T h e o t h e r p o i n t is a b o u t t h e r e p r e s e n t a t i o n of
g r a m m a t i c a l s t r u c t u r e s G r a m m a r d e s c r i p t i o n s
s h o u l d be e a s y to r e a d a n d write R e p r e s e n t a t i o n s by
u s i n g case f r a m e s m a k e r u l e s of a n y k i n d for N M P
v e r y simple, d e s c r i b i n g no N M P contents
I n o r d e r to d e a l w i t h t h e a b o v e t w o p o i n t s ,
c o m b i n i n g t h e c a s e a n a l y s i s w i t h A T N G - b a s e d
N M P ( V T Y P E - N M P ) s a r e d e a l t w i t h b y r e e u r s i v e
case-analyzing
2.3 Structured Constituent Buffer
A s mentioned above, syntactic and semantic
structures are basically derived from a sentence by
analyzing a C L A U S E Analysis control depends on
t h e c a s e f r a m e , w h e n t h e v e r b h a s b e e n j u s t
a p p e a r i n g in a C L A U S E H o w e v e r u n t i l s e e i n g the verb, all of t h e p h r a s e s , w h i c h m a y be n o u n p h r a s e s
w i t h e m b e d d e d clauses, P P s or A D V s before the verb,
m u s t be h e l d in c e r t a i n r e g i s t e r s or buffers
H e r e , a n e w b u f f e r , S T R u c t u r e d C O N s t i t u e n t
B u f f e r ( S T R C O N B ) , i s i n t r o d u c e d to h o l d t h e s e
p h r a s e s T h i s b u f f e r h a s s u r f a c e c o n s t i t u e n t s
s t r u c t u r e , a n d consists of specific slots T h e r e a r e two
s l o t t y p e s One is a r e g i s t e r to c o n t r o l E n g l i s h
a n a l y s i s a n d t h e o t h e r i s a b u f f e r to h o l d s o m e
m e n t i o n e d - a b o v e constituents The first type h a s two slots ; one is s i m i l a r to a b l a c k b o a r d a n d r e g i s t e r s t h e
n a m e s of unfilled-slots The o t h e r s t a c k s t h e n a m e s
of filled-slots in o r d e r of p h r a s e a p p e a r a n c e a n d is used for b a c k t r a c k i n g in t h e a n a l y s i s The second slot
t y p e i n v o l v e s s e v e r a l k i n d s of procedures One of the
m a i n p r o c e d u r e s , ~ g e t p h r a s e " , e x t r a c t s s o m e
c a n d i d a t e s for t h e slot f i l l e r from t h e l e f t side of a
C L A U S E I t fills t h e slot w i t h t h e s e c a n d i d a t e s T h i s procedure t a k e s one a r g u m e n t , which is a c o n s t i t u e n t
m a r k e r , ~ p r e p o s i t i o n a l - p h r a s e " , ~ n o u n - p h r a s e " a n d so
on (in practice, u s i n g each a b b r e v i a t i o n )
F o r e x a m p l e , w h e n t h e f o l l o w i n g s e n t e n c e is given, the e v a l u a t i o n for ~(getphrase 'preph)"in L I S P
r e t u r n s o n e s y m b o l g e n e r a t e d f o r t h e h e a d
p r e p o s i t i o n a l p h r a s e , ~ n the m a c h i n e l a n g u a g e " , and
d e t e r m i n e s t h e slot filler
(sl) '~In t h e m a c h i n e l a n g u a g e e a c h b a s i c
m a c h i n e o p e r a t i o n is r e p r e s e n t e d b y t h e
n u m e r i c a l code t h a t i n v o k e s i t in t h e computer, a n d "
However, if t h e a r g u m e n t is ~verb", t h i s procedure only tells t h a t the top word of unprocessed C L A U S E
is a verb A t t h a t m o m e n t , the process of f i l l i n g with
2.4 C L A U S E Analysis After seeing a verb in a C L A U S E , that is, filling the verb slot in the S T R C O N B , the case analysis starts W h e n the parser control moves on the case
first case slot, w h i c h is generally one for the
predicted in the slot for S U B J E C T in S T R C O N B Therfore, the predicted phrase is tested to determine
as a case instance The parser control moves to the next case slot and a candidate phrase for it is extracted from the remainder of the input sentence by
i n v o k i n g the function ~getphrase" w i t h N P -
Trang 3a r g u m e n t T h i s s l o t is u s u a l l y O B J E C T , or
obligatory prepositional phrase n a m e if the verb is
intransitive Furthermore, the control moves to the
n e x t case slot to fill i t , i f the case f r a m e h a s more
slots, all of which are obligatory case slots They are
described in a m e a n i n g s l o t (whose v a l u e is a
m e a n i n g frame) in a case frame, while optional case
slots are united in a special frame
The process to fill the case slots is continuing until
the end of the case frame Then, more t h a n one
candidate for a case s t r u c t u r e m a y be e x t r a c t e d
More than one for an N P extracted by " g e t p h r a s e "
gives m a n y case structures, because of the difference
in i n p u t remainders
Next, recusive p a r s i n g will be m e n t i o n e d In
a n a l y z i n g e m b e d d e d c l a u s e s , w h i c h are V T Y P E -
NMPs C L A U S E a n a l y s i s also gets in use of N P s
parsing It is supported with a new STRCONB The
procedure to call NP analysis is described in the n e x t
section The c o n c e p t u a l d i a g r a m for L U T E - E J
analysis as a recusive CLAUSE is shown in Fig.1
STRUCTURED-CONSTITUENT-BUFFER
l <*sub >
l
L Case Analysis ! ]
*case-frame*
<*agent>
<*object>
<*recipient >
STRUCTURED-CONSTITUENT-BUFFER •
L _ Case Analysis [
*case-frame*
~ STRUCTU~D-CONSTITUZNT-BUFFER I
Fig.1 Conceptual D i a g r a m of L U T E - E J Analysis
analysis of
i NOUN Phrase
ATNG-based analysis
process
(embedded clause,
noun clause
2.5 NP Analysis
An N'P structure is basically described as the rule
(R4) In this paper, NHD structure and the analysis
for it a r e o m i t t e d N M P is a n o t h e r m a i n N P
constituent and will be explained here
NM:P is described in the following form
(R5) < NMP > : : =
<PaSt-Participle-PH > / <ADJective-PH> /
<INFinitive-PH > / <RELative-PH > /
<CARDINAL> <UNIT> <ADJ>
If an NMP is represented by any kind of VP or ADJ-PH, it is described in a case structure by using a case frame That is, VTYPE-NMPs are parsed in the same way as CLAUSEs However, a VTYPE-NMP has one (or more) structural missing element (a hole)
c o m p a r e d w i t h a C L A U S E T h e r e f o r e , complementing them is needed by restoring a reduced form to the complete C L A U S E Extending " H O L D ' - manipulation in A T N m a k e s it possible This extension deals with not only relative clauses but also
V T Y P E - N M P s That is, the phrases with a "whiz- deletion" in Transformational G r a m m a r can be treated A D J - P H s can also be treated For example, the following phrase is discussed
(s2) '~I k n o w an actor suitable for the part."
In the above case, the deletion of the words, "who is",
results in the complete s e n t e n c e b e i n g t h e a b o v e representation The e x t e n d i n g HOLD-mm~ipulation holds the antecedent of a C L A U S E with a V T Y P E -
N M P Calling the case analysis recursively, the
V T Y P E - N M P is parsed by it Each V T Y P E - N M P has
a specific type, P R P - P H , P S P - P H , I N F - P H or A D J -
PH Each of them looks for an antecedent, as the object or the subject: so that each is treated according
to the procedure to decide the role of the antecedent and the omitting grammatical relation Therefore, it
is necessary to introduce one "context" representing
V T Y P E - N M P The present extension d e m a n d s the context with the antecedent a n d calls the case analysis
T h e following structured representation describes
a N O U N , as stated above
(NOUN (*TYPE ($value (instance))) (*CATEGORY ($value Csemantic-category'))}
(*SELF ($value ("entry-name'))) (*POS ($value (noun)))
(*MEANING ($value ("each-meaning-frame-list"))) (*NUMBER ($value ("singular-or-plural"))) (*MODIFIERS ($value CNHD-or-NMP-instance-list"))) (*MODIFYING ($value Cmodificand")))
(*APPOSITION($value (" appositional-phrase-instance"))) (*PRE ($value Cprepositional-phrase-instance"))) (*COORD ($value ("coordinate-phrase")))) Each word with prefix "*" describes a slot n a m e such
as a case f r a m e has H o w e v e r m a n y s l o t s a r e prepared for holding pointers to represent a syntactic structure of an NP The v a l u e for V T Y P E - N M P s
* M O D I F I E R S is a p a i r of V T Y P E - N M P s and an
i n d i v i d u a l v e r b a l symbol, for e x a m p l e , " ( P R P - P H verb*l)"
Trang 4Complementing NP's structure, an appositional
*APPOSITION-slot and treated in the same w a y as
another N M P by a pair of a delimiter ~," and a phrase
A Coordinate conjunction is another important
structure for an NP There are three kinds of
coordinates in the present N P rule The first is
between NPs, the second is N H D s , and the third is
NMPs The NP r e p r e s e n t a t i o n with t h a t conjunction
is described by an i n d i v i d u a l c o o r d i n a t e s t r u c t u r e
T h a t is, the conjunction looks like a predicate w i t h
a n y N P s a s p a r a m e t e r s , for e x a m p l e , ( a n d N P 1
NP2 NPi) Therfore, the coordinate s t r u c t u r e h a s
" * C O O R D I N A T E - O B J E C T S " and "*OBJ-CAT'" slot,
e a c h of w h i c h is f i l l e d w i t h a n y i n s t a n c i a t e d
N P / N H D / N M P s y m b o l or a n y c o o r d i n a t e t y p e ,
r e s p e c t i v e l y
Some l i n g u i s t i c h e u r i s t i c s a r e n e e d e d to p a r s e
NPs, along w i t h e x t r a c t i n g a s few i n a d e q u a t e N P
s t r u c t u r e s a s p o s s i b l e S e v e r a l h e u r i s t i c s a r e
i n t r o d u c e d into L U T E - E J parser T h e y are shown as
follows
(1) H e u r i s t i c s for a compound N P
" G e t p h r a s e " function v a l u e for an N P is the l i s t of
c a n d i d a t e s for a n a d e q u a t e N P s t r u c t u r e T h e
function first e x t r a c t s the l o n g e s t N P c a n d i d a t e from
a n input In t h i s a n a l y s i s , i t s end word is s e p a r a t e d
from t h e r e m a i n d e r of t h e i n p u t b y some heuristics,
(a) The top w o r d in t h e r e m a i n d e r i s a p e r s o n a l
pronoun
(b) Its end word h a s a p l u r a l form
(c) Its top is a d e t e r m i n e r
T h e s e h e u r i s t i c s p r e v e n t t h e v a l u e from h a v i n g
a b u n d a n t n o n - s e m a n t i c a l structures
(2) I-Ieuristics b y u s i n g contexts
W h e n N P a n a l y s i s is called when f i l l i n g a case
slot, the c a s e - m a r k e r ' s v a l u e for i t is d e l i v e r e d to N'P
a n a l y s i s T h i s v a l u e is c a l l e d " s y n t a c t i c l o c a l
context" I t is useful in rejecting pronouns, which are
u n g r s m m a t i c a l l y inflected, b y t e s t i n g the a g r e e m e n t
w i t h the s y n t a c t i c local context and t h e subject or the
object A n o t h e r c o n t e x t u s a g e is s h o w n b e l o w
A s s u m e t h a t a p h r a s e c o n t a i n i n g a c o o r d i n a t e
conjunction '~and", for example, is in a context which
is an object or a complement, and the word n e x t to the
c o n j u n c t i o n is a p r o n o u n I f t h e p r o n o u n is a
subjective case, the conjunction is d e t e r m i n e d to be
one b e t w e e n CLAUSEs To t h e c o n t r a r y , the pronoun
b e i n g a objective case d e t e r m i n e s the conjunction to
connect a n N P w i t h it
(3) Apposition
M a n y v a r i o u s k i n d s of a p p o s i t i o n s a r e u s e d in texts Most of t h e m are shown by N S a g e r [80] The preceding a p p o s i t i o n a l s t r u c t u r e s a r e used
3 L U T E - E J P a r s e r M e r i t s
In two sentences, each h a v i n g different s y n t a c t i c structures, t h e r e is a problem involved in i d e n t i f y i n g each case by e x t r a c t i n g s e m a n t i c r e l a t i o n s between a
p r e d i c a t e a n d a r g u m e n t s ( N P s , or N P s h a v i n g
p r e p o s i t i o n a l m a r k s ) L U T E - E J case a n a l y s i s h a s solved t h i s problem by i n t r o d u c i n g a new case slot
w i t h t h r e e components (Section 2.2.) F o r case frames
in L U T E - E J a n a l y s i s c o n t a i n i n g t h e s l o t s , a n
a n a l y s i s r e s u l t has two f e a t u r e s a t the s a m e t i m e One is a surface s y n t a c t i c s t r u c t u r e and the o t h e r is a
s e m a n t i c s t r u c t u r e in two slots Therefore, m a n y case
f r a m e s are p r e p a r e d according to p r e d i c a t e m e a n i n g s and case f r a m e s a r e p r e p a r e d according to p r e d i c a t e
on one predicate (verb)
A n analysis example is s h o w n for the s a m e semantic structure, according to which there are three different syntactic structures These three sentences are as follow (from Marcus [80] )
(s4) ~The judge presented the boy with the prize." (s5) "The judge presented the boy the prize." Three individual structures are obtained for each sentence and their meaning equivalence for each slot
by doing the same for case-names
Incidentally, a sentence containing another meaning of "present" is as follows It m e a n s "to show
In this case, the "present" frame m u s t prepare the obligatory "at" case slot
3.2 A n Effect of Combining Case Analysis with
A T N G - b a s e d Analysis The next section shows one application of the
L U T E - E J parser, which is a machine translation system So, taking the translated sample sentence in Section 4., effective points in parsing are s h o w n in this section The sample sentence is as follows
the instructions are complex statements, each equivalent to several m a c h i n e - l a n g u a g e instructions, a n d they refer to m e m o r y
O n e point is N M P analysis method by recursive calling for case frame analysis In the example, two
Trang 5NMP phrases are seen
modifies "each", appositive
to the preceding "statements",
and modifies "names"
These phrases are analyzed in the same case frame
a n a l y s i s , e x c e p t for t h e p h r a s e d e l e t i o n t y p e s
(depending on VTYPE-NMP) a p p e a r i n g in them The
deleted phrases are the subject p a r t and the object
p a r t respectively J u d g i n g from the point of a p a r s i n g
t r a n s p o r t s the deleted phrases, "each" and "names",
with the contexts to the case frame analysis
The other point is to hold undecided case elements
in STRCONB The head PP and the subject in the
sentences, for example, are buffering until seeing the
m a i n verb
4 An Application to Machine T r a n s l a t i o n
One of the effective applications can be shown by
c o n s i d e r i n g t h e N M P a n a l y s i s w i t h e m b e d d e d
phrases These NMPs are represented by instances of
actions, i.e i n d i v i d u a l case f r a m e s which m a y be
h a v i n g an unfilled case slot A p p l y i n g L U T E - E J
p a r s e r to an a u t o m a t i c machine t r a n s l a t i o n system,
there m a y be a little problem in l a c k i n g the case slots
i n f o r m a t i o n The r e a s o n is b e c a u s e t h e l a c k i n g information can be t h o u g h t of as b e i n g i n d i s p e n s a b l e for a semantic structure in one l a n g u a g e , for example
a t a r g e t l a n g u a g e J a p a n e s e , in spite of h a v i n g them
in a n o t h e r languages, for example a source l a n g u a g e English The p r o b l e m is t h e difference in how to modify a h e a d noun by an N M P or an e m b e d d e d clause
In J a p a n e s e , a N O U N is often m o d i f i e d by an embedded clause in the following pattern
"<predicate's arguments>* <predicate> NOUN"
; * representing recursive applications Therefore, in J a p a n e s e , an N M P p h r a s e represented
by a case frame corresponds to an embedded clause
a n d t h e v e r b of t h e f r a m e c o r r e s p o n d s to t h e predicate
A t r a n s l a t i o n example is shown in Fig.2
References
M a r c u s , M i t c h e l l P., "A T h e o r y o f S y n t a c t i c Recognition for N a t u r a l Language", MIT Press, 1980
S a g e r , N a o m i , " N a t u r a l L a n g u a g e I n f o r m a t i o n Processing", Addison-Wesley, 1981
Waltz, David L., "An E n g l i s h L a n g u a g e Q u e s t i o n -
A n s w e r i n g System for a L a n g u a g e R e l a t i o n a l D a t a Base", CACM Vol.21, 1978
Winograd, Terry, " L a n g u a g e as a Cognitive Process", Vol.1, Addison-Wesley, 1983
I l n t h e h ~ g h e r - l e u e l p r o g r - a m , ' ~ - - J ' ' ~ = ~ - - P ~ ; " - - ' - - " ~ "~ "-~'-='~- I
; n 9 l a n g u a g e s t h e i n s t r u c t [ o : I ' ~ ] / ' J £ / ' l ~ '~< J / / ~ I ~ C 3 •
n t S , e a c h e q u [ v a | e n t t o s e : A -~- ~ - ¢ r I
u C t [ O s ar~cl t h e ~ f e n c e r t o i ~/ "" {]'' " - - ~ I
l ~ , e m o r ~ t o c a t , o n s o ~ n a m e s c a t` ~ - - - - -
l O r i g i n a l T e x t ( E n g l i s h ) J - - - 4 ~ u ~ Z , .~Or~ ~ - • ~ = - = ~
I~ =~ ~ E ~ 4 T E I N L E : ~ ] t ; ~ E ] 2 E : C ~ t ~ D I D ~ T E ~L (
fr~Oi I U t ~ E= SEt 'TEt~CE : 0 8 1 8 E: CP4ND l D I ~ T E - 2 I I.,m,[ '~' E : P P E D I C ~ T E : e 8 2 4 E : U E R B = ~ I-'~-" ]-" ~-J'n~F[~_4' 75.~ Z ' 4] }~;F~'l'~'~ r"r ~
}t ~ [ l ( 1 t 0 _ E : E T ~ : 0 8 6 9 E : rlEIIORY l
I ( I t ' | ~ - : - - - E ' S ~ T E b ~ C C : O O t e E : C A , ' I D I D A T E 4 " " ~ "~" "-~ ' - ~ ~ ' ~ ' ' - -
I I~0L / : ~ ! £ ~ ELEMENT : 0 0 3 4 ~ ' C A S E - I~
I!!i I T!I !oii i =
I - - 16k~ ".pp'° ,.~,: ,T,~, ,ooo~- ' ,-,~T,,T,-,,-.T= - , j " ~ - " = ' '
_ " E : r O u H E A D E I ~ : (]~352 E ' A D J m 3 5
E : Q E "OO F EP 006.'2 E A D L ' E P B - 1 8 ~ , ~ i [ ~ ' ~
E ; i H E : P P E D I C A T ~ : k ~ Q 2 4 E ' A D J P H - 5 - ~ ~'4 ' ~ r h
E : EF4T R, ~ : {3869 E : EQU I UI~L Er tT " "
- - E : C~qSE - EL EMEt.IT : ~3054 E : C ~ S E - EL ErlEr4[ - 4 ~ a ~ ~m .~.,y, "1
G e n e r a t e d Internal Representation Processes Window
Fig 2 A n E x a m p l e of LUTE T r a n s l a t i o n Results on the Display
(from EngLish to J a p a n e s e )
158