Best ppt

1. MACHIME TRANSLATION
BY:SHANTANUARORA
BC2018049

2. Introduction
Main challenges
Techniques
Features

3. Introduction
Machine translation:-
Machine Translation has been defined as the process that
utilizes computer software to translate text from one
natural language(such as English) to another (such as
Arabic).
The idea of machine translation may be traced back to
the 17th century
MT on the web starts with Systran offering free translation
of small texts (1996)

4. Technique
Example-based MT
Dictionary-based
Rule-based
Hybrid MT
Neural MT
Statistical
Interlingual
Transfer-based

5. Example-based MT
characterized by its use of a bilingual corpus with parallel
texts as its main knowledge base.
It is essentially a translation by analogy
Ex
English
Arabic
English
Arabic
How much is that umbrella
‫هلظمال‬‫هذه‬‫رعس‬‫مك‬
How much is that doggie
‫ال‬‫ك‬‫ل‬‫ب‬‫اذه‬‫رعس‬‫مك‬

6. Dictionary-based
The words will be translated as a dictionary does — word
by word, usually without much correlation of meaning
between them

7. Rule-based
RBMT involves more information about the linguistics of the
source and target languages ,using the syntactic rules and
semantic analysis of both languages
This type of translation is used mostly in the creation
of dictionaries and grammar programs

8. Interlingual
instance of rule-based machine-translation
Itis necessary to have an intermediate
representation(interlingua) that captures the "meaning" of the
original sentence in order to generate the correct translation
"language neutral" representation that is independent of any
language
Advantage: one of the major advantages of this system is that the
interlingua becomes more valuable as the number of target
languages it can be turned into increases
the only interlingual machine translation system that has been
made operational at the commercial level is the KANT system

9. Transfer-based
Itis necessary to have an intermediate representation that
captures the "meaning" of the original sentence in order
to generate the correct translation
it depends partially on the language pair involved in the
translation

10. Statistical
using statistical methods based on bilingual text corpora,
such as the Canadian Hansard corpus
The idea behind statistical machine translation comes
from information theory

11. Hybrid MT
combination of statistical and rule-
based translation methodologies

12. Neural MT
neural network is trained by deep
learning techniques

13. Challenges in MT
Ambiguity
Ex1: Book the flight -> verb Read the book -> noun
Ex2: Kill a man (‫)لتق‬ Kill a process )‫(ءاهنا‬
Ex3:she couldn’t bear children
‫الفطالا‬‫لمحت‬‫ت‬‫س‬‫ت‬‫ط‬‫ي‬‫ع‬‫ال‬‫الفطا‬‫باجنا‬‫ت‬‫س‬‫ت‬‫ط‬‫ي‬‫ع‬‫ال‬

14. Challenges in MT
Different word orders
English word order : subject –verb –object
Mohamed is at home
Arabic word order:
) ‫هيمسا‬‫هلمج‬ ( ‫لزنمال‬‫يف‬‫ي‬‫ت‬‫و‬‫ا‬‫ج‬‫د‬‫دمحم‬
) ‫ف‬‫ع‬‫ل‬‫ي‬‫ه‬‫هلمج‬ ( ‫ال‬‫م‬‫ن‬‫ز‬‫ل‬‫يف‬‫دمحم‬‫جاوتي‬‫د‬
Japanese: subject –object- verb

15. Challenges in MT
Compound Words
Arabic
English
‫اه‬َ‫وم‬َ‫ك‬َ‫م‬َ‫ز‬َ‫ل‬َ‫ن‬ََ‫أ‬
Shall we compel you to accept it
Missing Names
A language may not have a word for a certain
action or object that exists in another language
ksnona (Greek)
guest room(english)

16. Application
Google translator
Bing Translator
SYSTRAN
Asia Online