Báo cáo khoa học: "An Evaluation of METAL: the LRC Machine Translation System" ppt

the LRC Machine T r a n s l a t i o n S y s t e m

Jonathan S l o c u m Microelectronics and C o ~ u t e r T e c h n o l o g y Corp

W i n f i e l d S Bennett

L e s l e y W h l f f i n

E d d a N o r c r o s s Siemens Co~mmmllcation Syst~-~q, Inc

A b s t r a C t

T h e Linguistics R e s e a r c h Center (LRC)

at the U n i v e r s i t y of Texas at A u s t i n is

evaluation, a n d will b r i e f l y indicate some

G e r m a n original tex~ and a raw (unedIt~3d,

translation of that text into English are

included as appendices

Introductlon The Linguistics R e s e a r c h Center (LRC)

at the U n i v e r s i t y of Texas at A u s t i n is

will b r i e f l y indicate some fur%Ire directions

reformatted) METAL translation of that

into English

H i s t o r y and Status Machine translation research at the

University of Texas b e g a n in 1956: the LRC

p r o v i d e d b y the U.S Air Force's R o m e A i r

Development Center and other U.S gove~,m~nt

agencies In 1979, Siemens A G b e g a n funding

the development p h a s e of the METAL m a c h i n e

i ~ l e m e n t a t l o n of the current system was

initiated A p r o t o t y p e has recently b e e n

delivered to the sponsor for m a r k e t testing

T h e current system is a unidirectional

G e r m a n - E n g l i s h system, although w o r k to add

ot/%er target languages, as well as c r e a t i n g

underway T h e p r e s e n t staff for the METAL

p r o j e c t consists of seven full-ti~e and five

A p p l i c a t i o n Environment Software h a s b e e n d e v e l o p e d to h a n d l e the formatting p r o b l e m s associated w i t h

in SNOBOL, a u t o m a t i c a l l y m a r k s and p r e p a r e s texts for the M E T A L translation s y s t e m

1984] T h e only h u m a n intervention p r i o r to translation is c h e c k i n g and c o r r e c t i n g the

routines P o s t e d i t i n g is e x p e c t e d for the output t ~ T h e s y s t e m does n o t expect (or p r o v i d e for) h u m a n intervention d u r i n g the actual t r a n s l a t i o n phase

P r e - p r o c e s s i n g a n d p o s t - e d l t i n g are

p r e s a n t l y done on a DEC-2060; the actual translation, on a Symbolics Lisp machine

T h e "proch/ction ~ " design envisions a

from w h i c h t,he p r e p a r e d ~ will b e sent

to the t r a n s l a t i o n u n i t and on w h i c h t h e output texts will b e postedlted

M E T A L uses a transfer a p p r o a c h for translation T h e entire p r o c e s s consists of

integration p h a s e works w i t h whole p a r s e

transfer and generation w e r e e s s e n t i a l l y a single phase, b u t w o r k is c u r r e n t l y u n d e r w a y

to separate this single p h a s e into ~wo, w i t h

a m u c h m o r e p o w e r f u l generation phase

L l n g u l s K i c Component

T h e curren~ M E T A L lexicon consists of over 20,000 G e r m a n and E n g l i s h m o n o l i n g u a l

and an a p p r o p r i a t e l y large number of transfer entries T h e featnlres and v a l u e s

and/or synthesis of these it~m~ d u r i n g the

are r e a s o n a b l y s 4 ~ l e , b u t entries for v e r b

p a r s e d b y word-level gr~m-~r rules, w i t h the stems and e~dlngs a s s i g n e d to appropriate lexical categories

E a c h t-Fans fer l e x i c a l e n t r y is a

F u r t h e r m o r e , t e s t s a n d o p e r a t i o n s ( i n c l u d i n g

t r a n s f e r en~rles

s e c t i o n for e a c h t a r g e t language: a t o p l l n e

o p t i o n a l e n u m e r a t i o n o f e a c h c o n s t i t u e n t ) ; a

w h i c h e n f o r c e a g r e e m e n t a m o n g t h e r i g h t - h a n d

t a r g e t - l a n g u a g e - d e p e n d e n t T r a n s f e r sections,

t r a n s l a t i o n into t h e t a r g e t language

a r e h a n d l e d u n i f o r m l y (i.e., w i t h o u t s p e c i a l

m e c h a n i s m

t/llngs as s u b j e c t - v e r b a g r e e m e n t

t r a n s f o r m a t i o n s m a y a l s o b e u s e d in t r a n s f e r

w r i t t e n as p a r t o f a r u l e or c a l l e d b y name

T h e l e x i c o n for M E T A L is m a i n t a i n e d v i a

w i t h o u t t h e a i d o f t h e I N T E R C 0 D E R or l e x i c a l

d e f a u l t e r

is d o n e b y m e a n s o f a n u m b e r o f m e n u - d r i v e n

a h i g h d e g r e e o f l e x i c o n i n t e g r i t y t h e M E T A L

T h e v a l i d a t l o n p r o c e s s is a u t o m a t i c a l l y u s e d

i n t r o d u c e d d u r i n g e d i t i n g

i n d i v i d u a l s o u r c e files

s u c c e s s f u l l y b u i l t o n e or m o r e Ss at a g i v e n

or m o r e S r e a d i n g s , t h e p a r s e r w i l l c o n t i n u e

M E T A L [Slocum e t al., 1984]

P o s t - E d l t o r s ' E v a l u a t i o n

p r o s p e c t u s (for a s p e e c h r e c o g n i t i o n system)

m ~ m h e r o f t h e M E T A L p r o j e c t , o n e not) T h i s

o f METAL) d u r i n g t h i s e x p e r i e n c e

63

#i 9hr 1 0 m l n 3hr 4 0 m i n 2br 1 0 m i n 1 0 9 5.5 4 4 1

(N.B T h e n u m b e r o f p a g e s o f t e x t w a s c o m p u t e d a u t o m a t i c a l l y o n t h e b a s i s o f

" S i e m e n s s t a n d a r d p a g e s " : 26 l i n e s x 55 c h a r a c t e r s = 1430 c h a r a c t e r s / p g )

T h e t a b l e a b o v e s u m m a r i z e s t h e editors'

b u t t h e i r t i m e s a r e r e l a t i v e l y close

C o m m e n t s b y E d i t o r #i :

t y p i n g errors

T h e ~ m p r e s s i o n o f p o s t - e d i t i n g w a s t h a t

in f o r m a t m a d e p o s t - e d l t i n g easier, a l t h o u g h

t h e r e is s t i l l r o o m for further e n h a n c e m e n t

lles i n t h e i n i t i a l p o s t - e d l t p h a s e w h i c h is

re- f o r m a t t i n g text

C o m m e n t s b y E d i t o r #2:

1984, t h e J u n e 8 4 o u t p u t s h o w e d c o n s i d e r a b l e

f u t u r e t r a n s l a t i o n s

F u t u r e D i r e c t i o n s

tex~c u s i n g a h a r d c o p y o f t h e M E T A L o u t p u t

u p o n w h i c h p r e l i m i n a r y p o s t - e d l t i n g h a s b e e n

p e r f o r m e d it w a s e a s i e r to c o n c e n t r a t e o n

t e s t s b u t o n e w a s b e t t e r a b l e to d i s t i n g u i s h

p r o c e s s w a s less t i r i n g t h a n in t h e past

o u t p u t h a s i ~ 3 r o v e d t h e r e are s t i l l [some]

r e f o r m a t t i n g p r o g r a m ]

r e l e a s e d for m a r k e t t e s t i n g i n J a n u a r y 1985

o t h e r t a r g e t languages, s p e c i f i c a l l y S p a n i s h

syst~m~

R e f e r e n c e s Slocum, J., a n d W S Bennett, "The L R C

M a c h i n e T r a n s l a t i o n System: A n A p p l i c a t i o n

o f S ~ a t e - o f - t h e - A r t T e x t a n d N a t u r a l

L a n g u a g e P r o c e s s i n g T e c h n i q u e s to t h e

T r a n s l a t i o n o f T e c h n i c a l Manuals, " W o r k i n g

P a p e r L R C - 8 2 - 1 , L i n g u i s t i c s R e s e a r c h Center,

U n i v e r s i t y o f Texas, J u l y 1982

Slocum, J., e t al., "METAL: T h e L R C M a c h i n e

T r a n s l a t i o n System, " p r e s e n t e d at t h e I S S C O

T u t o r i a l o n M a c h i n e T r a n s l a t i o n , Lugano,

as W o r k i n g P a p e r L R C - 8 4 - 2 , L i n g u i s t i c s

R e s e a r c h Center, U n i v e r s i T y o f Texas, A p r i l

1984

ooj !

-4

U

£

U ~ J , : ~ I ~ ",'* Q, q) U ~ g @ B ~J @ , ' ~ g O O L I , O ~ O u t ~ U " Q , O .,.I , - ~ U

• ,,~ g , * m ,.~ s~ ~1 g ~ , L U g e g " ~ ' ~ e

65

B

4J ~.,,4 ,,-¢

° = ° ~ ' ~ ° " " " ° ° ° - - - : o : o o " )~i , ' ~ , ' o = ~ ~ ;

> (0 "0 ,-~ ~ • U I, ' - ~l r ' 0 0 0 ~ >

, , , , N - , ~, -, ',,., U u U ~ ' ~ " ,w 0 ~

U ~ U O U O U k : ~ m ~

• ~ 0 ~ ~ = ~ U

o > , ~ , ~ , § ~ u , ~ ' ® ~ ® o ~ ' ~ ® ~

• ~ 0 ,.I " > U g 0 ,-'i ','~ *~ ',~ 3 c -,'( "0 C:: 'l~ ~ u

U "~ • ,C := ea ~ "0 ~ (g m " l • U " 0 r" Q ~ Q U > '~

g ~l ~) > ' 0 i 0 @ ,-I (I) O ~ ,-i.kJ ~ ~ Vl ~ m , G G ~ C (~', , , C ~ , C ~ , O , C ; ~J O.-.i ~ r , m m •

C a ~ q , ~

0 C O B

C

~ ' " g ; o o O ~

U t

~ U N

,,,i ,,-i U

i~ °

,.,= ~ ~ ~ & , ~ =

e,

m

! ' ° - o.o

~ ' m w "o i ~ ® z o~, , ,~ l a: ® , a ~ ~ = c = c i ~ i c ~ ' o o~ i o ~ ~ , ' o o

i ° i ° " ° ° " ~ ° ° " ° " ' ° + ~ ~ °

° ° : < ' " ' " " ° ° I ~ i i o o , o o , o , , , o ~ o o ~ o o ,~

.in "0 r" ~ ~ ~ 'T g ~ , l ~- ~; -&:; [ ~ ' U t : n ,.CU • • V d -~ : 31 ! ~ ' 0 ~J , • r ' , Q r" " ' 0 l • := I~ ~ g ~ i n ~ , ~ 0 u O

U ' O ~ ~ ( 4 e U r , E ~ , O ~ (I) ~ O O :3 g g "0 Q O) i W O,-.I

4,.I ~ g * - 4 ' , g r" g ' ~ N ~ g ¢ ) ~ - e , , - ,,,-, ,w

, ' ~ C > ~ o r *

::3 ~ ' g U

O O U r O

r" ~ I~ 0 1 ~

'0

, ; ~ 5 = =

l J o o

- > o

~ .,., , ® ® = ~ ®

~ c" , ~ ', 4~ 0 ~1 " a > 0 0

I/)

F " " § ~ ® o , " , ~ o , "

, ~ : ~ - , ~ , - >,,, = ® ® , , = 5 o ® ~ o , ~

Q , ~ : ~ m E- , ~ L U O ~ o ~ e , t C , i m U , * C "~ > ~ ~ e ¢ = • I- u

,~ ~ , - = , , = , - , - , = , , , o , ' = ) - , ® = , ' o , , = , , , ' 5 = - ~ ~' ~ ~ = " ;

~ 1 U g g ' ~ ¢@ 0 O ~ ; " ' I t ~ O 4 J ~-t L "(:1(11 0 ~ "1 ~ (~ ~J r" ~ "1::1

; = o - 4 ® J'~.-,'~ , -, " 8 , - , - ~ o , , - o g ® ='~=> o ~ o

b - ~ ~ , 0 C ,,C U > ~ t - , C = U',-* L g m ~ cn ,'~U

5

g

~ ~ ~ 0 "" ~ ~

~ , = = = , , = ~ = = = = = = = = = = = = = = = = = = = = = = = = .= , , , , = ,., - o - ,

" ° " ' " " " ° ° ° o o o o o

O C O ~ ~ > , ~ ' , ~ > ~ ~ @ ~ U e" ~ ~ 0 ~ ~ , ~ 0 O 0

L

~ - U O

J ~ O g a ~ 4 J O ~ J 0 - = ~ 0 2 ~ " ~ : ~ ~ J O U ~ & I ~ l i ~ 0 ~ " ~ C C ~ - q O 0 g L O @ ~ ' ~ e ¢ U , '-¢ '~ C ~ .,-I e ' ' O • ~I,~ >~ O ~ U

U ~ > U ~

~ ~ ~ ® ~ : , ~ ®, ,,., ,', ~ ® ~ ~ ~ c " P , ~ ~ e

~ ~J ~, r ' G C r" rA ~ 3 B ' O ,Q ~ ~ ,~ *- a l ~ S ~ , - * ~ 0 ~ ~ - " 0 r" e" • C O G.,Q • " ",'* U -* m-,-* ,-~ b.i C ~ "."~ •

J z

O ~ , ~ U U r" ~ " E I -u o

C-,'* • >.,'~ ~ g e e ~ J ~- n-" ~ ' 0

,, ~ , ~ ~ = = ' o - ~ ' o u o, o o

.U ,.~ ~ , e ~ 0 ~'* ~ 0 ~ 1

U ~- 0 ¢1 ,,.< I, U ~ ~, C -'~ • ~ -'~ O ~ ) > C C ~ e , f , B M L ,,=i ,-I ~, ,.,i ~ g " ~J O~4J ~ ,,.i

O g ~ C U g O g • C i ~ '~ " I,) J~ R ~ 0 ,,.,~ I U ,-~ ,-~ t., ~1 ," r~ ~ , I g -,I ~ ~,~ U

~ , u ~ ~ ~ ~.,-.~ ~ ~J g ~ ~ ' C I J~ ~J C • U .~ -i I ~ n - O 0 L ~:~ 0 ~ L- ~

"O ~ Q O Q Q , r" O ~ g g ~ ZJ ~ Q , " O ' ~ ~ | ~J ~J ~ U > ' U J~

< ~ ~ ~ - ~ , > - ~ ~

,,,r= ~" C ' ~ ~11 ~J ,-~ ~ U ~ O" O ~ O

m ~ .~ ~ g ~ ~=.,.~ r ' ~ ~ n g 0 , , ~

¢q U t~ -"1J~ g f,,, 'El t" ¢1 g 4J g J~ I~ g },,, I,, N 'O 4J P' c" g } , ~ r ' O I ~ 1 ~ .~ - i i,,, e,, U

• ,=1 ,.< ~ g

O g U

69