A Linguistic Approach for Multilingual Machine Translation System

1
A Linguistic Approach for Multilingual Machine
Translation System
R2I 11

• Fahima Bouzit Mohamed Tayeb Laskri

Rencontres sur la Recherche en Informatique june
12-14, 2011
2
PLAN

• Introduction
• Analysis Levels
• Machine Translation
• Proposed Approach
• Fillmore Theory
• Conceptual Dependency
• Semantic Traits of Chafe
• Frame Based Representation
• Conclusion Perspectives

3
Introduction

• Language
• Natural Language Processing(NLP)

4
Introduction

• NLP Schools

• Linguistic Approaches
• Probabilistic (Statistical) Approaches

5
Introduction Analysis Levels

• Analysis Levels

• Morphology
• Syntax
• Semantic
• Pragmatic

6
Introduction Analysis Levels Machine Translation

• Machine Translation
• Translation
• Source language
• Target language
• Machine

7
Introduction Analysis Levels Machine Translation
The challenge in machine translation how to
program a computer that will "understand" a text
as a person does, and that will "create" a new
text in the target language that "sounds" as if
it has been written by a person. This problem may
be approached in a number of ways.
8
Introduction Analysis Levels Machine Translation
Basic Model of a Machine Translation System
9
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
French sentence
Arabic Sentence

Analyse
Construction
Frame in Arabic
Frame in French
Translation
Proposed Architecture
10
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• Proposed Approach
• Fillmore theory
• Conceptual Dependency (Schank)
• Nouns Classification (Chafe)
• Frame based representation (Minsky)

11
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Fillmore theory

• The sentence
• Verb Kernel
• Other components of the sentence peripherals
• Verbs typological nature

12
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• The case AGENT  syntactic case Subject.
•  
• The case OBJET  syntactic case objectComp
• Or
• syntactic case Subject
• verb mode Passive
•  
• The case INSTRUMENT  gram case Dative
• Preposition ??????, ???????? , ?
  ?
•  

13
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• The case SOURCE  grammatical case Dative
• Preposition
  ????
• Or
• A place noun playing
  the role
• of a direct object
  comp of
• some known verbs,
  such us 
• ????, ???like in
  ???? ????? ??????

14
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• DESTINATION  gram case Dative
• Preposition ? ?? ,
  ???? , ???? ,
• 
  ?????, ??????
  Or
• A place noun playing
  the role
• of a direct object comp
  of
• some known verbs, such
  us
• ??? in ??? ???????
  ?????? ?

15
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• FURNISHER  syntactic case Indirect object
• 
  complement.
• Animation Animated
• kind of verb transfert
  verb
• Particule ???? or ???
  ???
• eg ????? in ?????
  ??????? ?? ????? ?????

16

• BENEFICIARY  Syn case Direct object comp
• Animation Animated
• Kind of verb verb of
• transfert such us ???? ,
  ????? , ?????,
• , ????,
  ??????, ?????
• Particule ??? ? ???
• like ??????? in
  
• ???? ????? ???????
  ??????????? ??? ???????
  
• Or
• Syn case indirect object
  comp
• Animation Animated
• Kind of verb transfert
  verb
• like  ???? in ????
  ???? ???? ?????

17
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Traits of Chafe
????????
Animated,
Human,
Feminine,
()
()
(-)
Concrete,
Potent,
()
Unique,
()
(-)

Countable
()
??????
Animated,
Human,
Feminine,
(-)
(-)
()
Unique,
Concrete,
Potent,
()
()
(-)

Countable
()
18
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Frames
Fig 1. General Frame
19
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Frames
Fig 2. Specialized Frame
20
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Conceptual Dependency
PROPEL Apply a force to something
MOVE Moving a body part
GRASP Catch an object
INGEST Ingest, for a moving object
EXPEL Physically expel, for a moving object
PTRANS Move a physical object
ATRANS Modify an abstract relationship, such as possession
SPEAK Produce a sound support of an action such as Communicate
ATTEND Apply his attention to a perception or stimulus
MTRANS Information Transfer
MBUILD Creating a new though
21
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
French Sentence
Arabic Sentence

Analyse
Construction
Frame in Arabic
Frame in French
Translation
Proposed Architecture
22
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• Construction French
• Affirmative Sentence
• French Subject Verb Object deSource
• aDestination
• English Subject Verb Object fromSource
• toDestination
• Negative Sentence
• French Subject ne (n) verb pas object
• English Subject auxiliary not strem
• Imperative Sentence verb(infinitive) object

23
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• Re-organization
• French
• Le livre est vendu / beau
• La revue est vendue / belle
• Les livres sont vendus / beaux
• Les revues sont vendues / belles
• Less in English

24
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Examples

• ???? ???????? ????? ?????
• ??? ????? ???? ?????????
• ??? ????? ???? ?????
• ???? ???? ????????
• ??? ??????? ????? ?????????
• ???? ????? ????? ?????????? ??? ???????
• ???? ????? ????? ????????? ??? ??????

25
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives

• The user re-named the file
• The Child printed the text with the printer
• The Child printed text
• The Printed text printer
• The Engineer copied the database
• The child sended an email to the teacher
• The child sended an email to his teacher

• L'utilisateur a re-nommé le fichier
• Lenfant a imprimé le texte avec limprimante
• Enfant a imprimé le texte rapidement
• Le texte a été imprimé
• Lingénieur a copié la base de données
• Lenfant a envoyé un email à l'enseignant
• Envoyer un email à son enseignant

26
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Examples

• ???? ???????? ????? ?????
• ??? ????? ???? ?????????
• ??? ????? ???? ?????
• ???? ???? ????????
• ??? ??????? ????? ?????????
• ???? ????? ????? ?????????? ??? ???????
• ???? ????? ????? ????????? ??? ??????

• L'utilisateur de renommer le fichier
• Imprimante enfant du texte imprimé
• Enfant texte imprimé rapidement
• Imprimé imprimante texte
• Ingénieur de base de données de copie
• Envoyer un email à l'enfant de l'enseignant
• Envoyer un email à l'enfant mentor

27
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Examples

• ???? ???????? ????? ?????
• ??? ????? ???? ?????????
• ??? ????? ???? ?????
• ???? ???? ????????
• ??? ??????? ????? ?????????
• ???? ????? ????? ?????????? ??? ???????
• ???? ????? ????? ????????? ??? ??????

• User re-naming the file
• Child printer printed text
• Child printed text quickly
• Printed text printer
• Copy Database Engineer
• Send an email to the child the teacher
• Send an email to the child mentor

28
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Our translation system that some modules were
exposed in this paper, takes part in the semantic
processing of texts using purely linguistic tools
and finds fulfillment with the DCF method as a
basis.
This method has
been proved appropriate to the Arabic language
and its particularities as to syntax and semantic
sides 346
29
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Multilingual system
Enrich dictionaries to cover other domains
Fusion of linguistic and probabilistic approaches
30
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Enrich dictionaries to cover other domains
31
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Multilingual system
31
32
Introduction Analysis Levels Machine Translation Proposed Approach Conclusion perspectives
Fusion of linguistic and probabilistic approaches
33
Thank You !