Evolution of modern machine translators: Difference between revisions

From Computer Science Wiki
No edit summary
 
(3 intermediate revisions by the same user not shown)
Line 5: Line 5:
On a basic level, MT performs simple substitution of words in one language for words in another, but that alone usually cannot produce a good translation of a text because recognition of whole phrases and their closest counterparts in the target language is needed. <ref>https://en.wikipedia.org/wiki/Machine_translation</ref>
On a basic level, MT performs simple substitution of words in one language for words in another, but that alone usually cannot produce a good translation of a text because recognition of whole phrases and their closest counterparts in the target language is needed. <ref>https://en.wikipedia.org/wiki/Machine_translation</ref>


== Very good link ==  
== Evolution of machine translators ==  


* [https://medium.freecodecamp.org/a-history-of-machine-translation-from-the-cold-war-to-deep-learning-f1d335ce8b5 Please carefully study this resource. It is superb resource for the history of machine translation]. You should know the terms below at an '''outline''' level. Please do not use the abbreviations in the linked article in any IB answer.
* [https://medium.freecodecamp.org/a-history-of-machine-translation-from-the-cold-war-to-deep-learning-f1d335ce8b5 Please carefully study this resource. It is superb resource for the history of machine translation]. You should know the terms below at an '''outline''' level. Please do not use the abbreviations in the linked article in any IB answer.


# Rule-based machine translation
* Rule-based machine translation
# Example-based Machine Translation
uses dictionaries and a set of linguistic rules to translate between two languages. <ref>Sheevankit the awesome</ref>
# Statistical Machine Translation
* Example-based Machine Translation
# Neural Machine Translation  
uses ready-made phrases as examples instead of repeated translation. <ref>Sheevankit the awesome</ref>
* Statistical Machine Translation
the machine tries to recognise patterns by studying similar texts without the need for dictionaries or rules. <ref>Sheevankit the awesome</ref>
* Neural Machine Translation
uses a large artificial neural network to predict the probability of a sequence of words.<ref>Sheevankit the awesome</ref>


== Our ultimate goal ==  
== Our ultimate goal ==  

Latest revision as of 09:53, 27 February 2019

HL content: Modeling & Simulation[1]

Machine translation, (sometimes referred to by the abbreviation MT) is a sub-field of computational linguistics that investigates the use of software to translate text or speech from one language to another.

On a basic level, MT performs simple substitution of words in one language for words in another, but that alone usually cannot produce a good translation of a text because recognition of whole phrases and their closest counterparts in the target language is needed. [2]

Evolution of machine translators[edit]

  • Rule-based machine translation
uses dictionaries and a set of linguistic rules to translate between two languages. [3]
  • Example-based Machine Translation
uses ready-made phrases as examples instead of repeated translation. [4]
  • Statistical Machine Translation
the machine tries to recognise patterns by studying similar texts without the need for dictionaries or rules. [5]
  • Neural Machine Translation
uses a large artificial neural network to predict the probability of a sequence of words.[6]

Our ultimate goal[edit]

Instant, perfectly accurate translation. Click here for a funny take on this

Standards[edit]

  • Outline the evolution of modern machine translators.

References[edit]

  1. http://www.flaticon.com/
  2. https://en.wikipedia.org/wiki/Machine_translation
  3. Sheevankit the awesome
  4. Sheevankit the awesome
  5. Sheevankit the awesome
  6. Sheevankit the awesome