Translation studiesLesson 2 of 3 in the course Localisation and technology

Machine Translation: Approaches, Challenges, Uses, and Evaluation

5 flashcards · 5 quiz questions · 5 match pairs · 5 fill-in sentences · English

Save course
Lesson progress
0 / 5

Your progress is kept in this browser only.Sign up to save it

Summary

Machine translation (MT) uses computers to translate text or speech between languages. This lesson introduces what MT aims to do and traces important milestones from early ideas to web-based translation.

Machine translation is the use of computational techniques to translate text or speech from one language into another. A useful translation needs to convey more than individual word meanings: it should also handle grammar, idioms, tone, context, and pragmatic meaning.

All flashcards

Why can technical documentation be suitable for MT customization?
It often uses consistent terminology and formal language, reducing variation in the translation task.
What is a named entity?
A specific real-world person, organization, company, place, date, or quantity.
What was the significance of AltaVista Babel Fish?
It made SYSTRAN web translation available in 1997.
What is neural machine translation (NMT)?
Machine translation that uses deep-learning models to learn patterns from language data.
Why can human post-editing be useful after NMT?
It can help ensure that fluent output preserves meaning and suits its intended use.

Quiz questions

  1. Which statement best distinguishes interlingual MT from transfer-based MT?Answer: Interlingual MT uses a language-neutral representation; transfer-based MT uses one tied partly to a language pair.
  2. What kind of material does statistical machine translation learn translation patterns from?Answer: Bilingual parallel corpora containing texts and their translations.
  3. What does transliteration do when handling a name?Answer: Represents the name using corresponding letters in another writing system.
  4. What is an important limitation of automated translation metrics such as BLEU?Answer: They enable standardized comparisons but cannot fully measure communication in context.
  5. Why does machine translation require special caution in medicine and law?Answer: Errors can have serious consequences, and specialized meanings may be precise.

Based on the Wikipedia article Machine translation, licensed CC BY-SA 4.0. Adapted into a summary, flashcards and quizzes.

More in Translation studies

All Translation studies lessons
Translation studiesLocalisation Across Markets: Adaptation, Workflows, and Language Technologies5 flashcards · 5 quiz questions · 4 gamesPlay free →Translation studiesComputer-Assisted Translation Tools: Memory, Search, Alignment, and Augmented Translation5 flashcards · 5 quiz questions · 4 gamesPlay free →