ParaMT:
A Paraphraser for Machine Translation
! " # $ % &
! " # $ % &
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` a bca d P Q Oe S
! " # $ % &
! " # $ % &
•
•
•
•
http://www.nooj4nlp.net/
•
•
http://www.linguateca.pt/COMPARA)
•
http://www.linguateca.pt/metra)
•
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
mesa,N+FLX=CASA+CO+surf+EN=table
cair,V+FLX=ATRAIR+INMO+IntoType+EN=fall
holandês,A+FLX=INGLÊS+AN+lang+EN=Dutch
actualmente,ADV+FLX=FACILMENTE+TEMP+punc+pres+EN=nowadays
alguém,PRO+IMPERS+INDEF+EN=somebody
porque,RELINT+why+EN=why
e,CONJ+JOIN+EN=and
durante,PREP+TEMP+EN=during
cada,DET+IMPERS+INDEF+SG+EN=each
terceiro+NUM+ord+EN=one third
a curto prazo,ADV+TEMP+EN=in the short run
a favor de,PREP+CAUS+EN=in favor of
cada um,PRO+INDEF+SG+EN=each one
de quem,INT+ThatType+EN=whose
quem quer que seja,REL+WhateverType+EN=whoever
além disso,CONJ+COOR+EN=besides
um quarto,NUM+frac+EN=one fourth
] ^ _` ^ ] a b _c d ] e f g hij k lm fn fo g pi g c qr s g m f kr t u _r t v ] _ ^ r ] w ` ^ c x ]y y` u _ ] e f g hij k lm fn fo g zo{ ` ` s g m f k w _` ` |y { a r } w c wa ^ ] ] s r` ~sa r ] e f g hij k z g l g sa u g m f k ] s r` t ` sa r ^ _ a q } g f l w c wa ^ ] ] s r` ~sa r ] e f g hij k z g l g sa u g m f k w ` ` c g y s ] q b r` _ ^ c s ] _ ] q d ] e g f g _c ^ g m f k ` _ ] q bc yu s ] c | _ a r ]`q e g hij k i o g f g ^ c y g m f k ` u { t | c _ a r ] q ]r u _ { ` _ ]` e g i ` r a | c g m l p g m f k a q { t c y t ` _ { _u q ` _ ] ^ c yc _ x a ` e g g t _ g m f k ` u { ` ` _ ^ c _ t |u a { ` { c |` e g i ` r a | c g m l p g u qr ]y { g m f k ] s ` q b { a | c ] b` a y { ` e zo f g zo o g m f k a c d q ` e zo f g zo o g m f k`q s ` q bc _ | c y `| ]y y a | e zo f g g m f k c c x qy` d u q { ] c| q { c r` | e p m p g o z g m f k ] s ` qb a { t ^a _c a { ] ^ c e p m p g i ` r g g m f k ] { { t c _ a b t { ` c |r` q ` _ | a ^ ] ^ c r` | e p m p g io g m f k a qr` q b _uc qrc a { tGeneral dictionary
sample representing all
PoS, variable and
invariable forms
Sample of the
dictionary of Terms
and
Multiword Expressions
DicTUM
Sample of invariable
compounds in the
general dictionary
Sample of the
dictionary of
Biomedical Terms
Sample of the
dictionary of
Proper Names
¡ ¢ ¢ £ ¢ ¤ ¥ £ ¢ ¤ § ¨ © ª « ¬ ¢ Representation abstract language
Hierarchical taxonomy (sets, supersets and (sometimes) subsets)
Based on Logos SAL ontology
Integrated in the dictionary
It represents both meaning (semantics), and structure (syntax)
Over 1,000 categories
¡ ¢ ¢ £ ¢ ¤ ¥ £ ¢ ¤ § ¨ © ª « ¬ ¢ oun Supersets
concrete
mass
animate
place
information
abstract
process (intr)
process (tr)
measure
time
aspective
Sets
and Subsets of the
CONCRETE Noun Superset
Click on
CONCRETE Superset,
sets
and
subsets
for explanations
functionals
receptacles
bearing surfaces
links/bridges
thresholds, focal
points, barriers
conduits
fasteners
devices, tools
cloth thing
structural elements
concretizations of
verbals
concretizations of
mass nouns
undifferentiated
functionals
product/brand
names
* * *
agentives
software
vehicles
meters
machines/systems
communication agents
concrete chemical
agents
undifferentiated
agentives
* * *
natural things
minute flora
plants
trees
trees/wood
miscellaneous natural
things
* * *
other concrete sets
*
impulses/lights
blemishes/marks
edibles (non-mass)
edibles/color
classifiers
amorphous
atomistic
undifferentiated
concrete things
* * *
*With one exception, these
sets
have no subsets
! " # " # $ % & ' $ % () * + , - . () + , - / 0 1, 2 3 . , () 4 $ 4' 4 5 6 4 7 4 5 6 2 / + / 8 9: + 0 ; / ( 3 . / < = 2 , < ): * 81> ( , 2 ) 6 4 5 ? @ 6 6 4 5 A / ++ . (9 0 2 /B . ( / B = C * , - / 1 ) + ) D ( = 1, 2 / 5 4 7 5 4 + ( *2 E 0 : 3, F / * + ) B G H . 8 6 ' 6 ' 2 8 ) 2 E 0 D / * D . B /-G B . + () 4 $ 6 6 4 $ 4 $ 6 6 ( / <, ) 0 ( / < / ( ( = <, ) % 4 $ 7 ! % 4 + ( , -E . + 0 ) (- /B .-+ ) (- / B .-+ ) ! 4 ? $ $ 7 $ $ 7 F 8, . ( : / 8, 2 / + . % ' % - / ,8 0 + .-< ) - F ( . D) # ' ' % 4 ' % + / A8 . 0 : 3 . 8 1 B : . / ' 4 I 4 7 ' 4I A ) ++ 8 . 0 A / (( . 8 D / (( / 1 / 4 $ ! 4 $ ! 2 3 * + . 0 / ( + . (9 / ( + J ( , / 5 ' 5 $ 7 ! ?% $ 4 7 I$ ? # '' ' # '' ; / 88 0 < ) ) ( F ) ( + / 7 K ? # ' ! 7 K 2 , (2 * , + 0 - . (H . 2 , (2 * , + ) 4 7 $ 5 5 4 7 $ 5 : 3, ( + 0 A8 /-E . + 2 / B , : ) 8 / ' 4 ' 7 7 6 ' 4 : F / ( 0 A ) - . ):: ) 4 $ 4' L $ $ % ' # 7 ' # + 3 ( . / <, - D 4 $ 4' L $ $ % 6 $ 6 /2 , < 8, -, - D I '$ ! 4 ? # ' ! 6 # ' ! M , -< ) : ; NO M , -< ) : ; NO ' 7 5 " # " # 6 %7 $ ' %7 $ ' / 8 D / . 0 : F ) ( . / 8 D / I 7 I 7 (): . 0 ; . . < . (H / ' ' /F F 8 . 0 ; ,88 ) ; B /2 , . , ( / ' ? &$ $ ! ' & ! ) / E 0 B /F 8 . 2 / (H / 8 3 ) 6 4 P ' 7 5 6 F . A A8 . 0 , 2 . A . ( D , 2 . A . ( D ! #7 Q $ R 6 S ! 6 F ) ( E 2 3 ) F 2 ): + ) 8 . + / ! #7 ? 4 $ 7 $ ' ! 4 $ 7 ) ( /- D . 0 2 3 . ((9 8 / ( /-T / 6 I 7 ?7 5 $ 7 5 8 /B F 0 A . / B 8U B F / < / #7 6 5 ? 6 ' K #7 6 : 2 ( / + 2 3 0 1 ( .2 E8 . : / ( < / 4 7 % ' 4 7 . 8 .B .-+ . 8 .B .-+ ) 6 $ ' I 5 $ 6 $ ' A ( . .V . 0 + , < . A ( , : / $ 6 4 $ 6 . 8 .2 + () -0 / + ) B = + ) B ) ! %% ' ! $ #W + ( , 18 . 0 2 * ( , ) $ ' $ % X YZ Z [ $
! " # $ % & '( " ) $ '*+ & & '" , , - . / - 0 '* & $ 1 0 ' ! , 2 2 1 $ & ( 3 &+ " 4 , 2 2 ( 3 %% 0 ' $ % " &+ , ! & ( +
& & )* &
% ) ) , . 5 , 6 3 " " $ 0 + )& 1 0 6&+ 7 3 + & ' & " $ 8 & % 9 ' , 0 + $ ' " & $ # & 3 + : , 0&+ )" & ! $ 0 $ 1 0 ' , 5 2 2 ! 0 %% 3 + $ & + 0 $ 7 0 + ' " , 5 '* $ # 3 + ! $ % & . , , , ! & ( + & & $ ( + 0 ) ) $ ) 3 + & $ ' $ % ; Z < "= YZ $
X $ %7 4 $ 7 ' ! 6 < !W 4 %7 4 $ 7 ' ! 6 '$ $ %7 4 $ 7 ' ! 6 ' 4 $ ' 6 < ! ' # %7 4 $ 7 ' ! 6 ' 6 ' %7 4 $ 7 ' ! 6 ''$ '$ $ %7 4 $ 7 ' ! 6 X $ 6 7 L $ ' $ 7 ' ! 6
Translation and
bilingual paraphrasing
of simple sentences
Graph to translate simple
sentences
NDRV
ADRV
VSUP
VCOP
AVDRV
Npred
Recognition and monolingual paraphrasing
of support verb constructions
(support verb construction / morphologically related lexical verb)
Recognition and paraphrasing of elementary
support verb constructions
co-occurring with predicate nouns
of the biomedical field
(support verb construction / lexical verb or
stylistic variant / non-elementary support verb
construction)
realizar efectuar
Interactive ReWriter
for word processing applications
such as text editing
Recognition and bilingual paraphrasing of support verb constructions
(Portuguese support verb construction / corresponding English verb)
SVC Recognition
Precision
SVC Recognition
Recall
SVC Paraphrasing
Precision
Pôr
73/73 - 100%
73/100 – 73%
72/73 - 98.6%
Tomar
75/75 - 100%
75/100 – 75%
68/73 - 93.1%
Ter
65/65 - 100%
65/100 – 65%
59/65 - 90.7%
Dar
57/60 - 95%
57/100 – 57%
46/51 - 90.1%
Fazer
43/45 – 95.5%
43/100 – 43%
40/45 - 88.8%
Average
62.6/63.6 - 98.4%
62.6/100 - 62.6%
57/61 - 93.4%
Evaluation of recognition and paraphrasing
of support verb constructions
Fundação para a Ciência e a Tecnologia
Fundação para a Computação Científica
Nacional
ParaMT:
A Paraphraser for Machine Translation