Evolution of modern machine translators

HL content: Modeling & Simulation

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.

Evolution of machine translators

• 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.

1. Rule-based machine translation

uses dictionaries and a set of linguistic rules to translate between two languages. 

1. Example-based Machine Translation

uses ready-made phrases as examples instead of repeated translation. 

1. Statistical Machine Translation

the machine tries to recognise patterns by studying similar texts without the need for dictionaries or rules. 

1. Neural Machine Translation

uses a large artificial neural network to predict the probability of a sequence of words.

Our ultimate goal

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

Standards

• Outline the evolution of modern machine translators.

Give a brief account.

a group or system of interconnected people or things.

Give an expected result of an upcoming action or event.

Give a brief account.