MTInClassHandout - L INGUISTICS 384: LANGUAGE AND COMPUTERS...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: L INGUISTICS 384: LANGUAGE AND COMPUTERS MACHINE TRANSLATION: IN-CLASS HAND-OUT 1 Introduction Machine translation (MT) systems translate automatically from one natural (hu- man) language into another. Since the 1950s, it has been an active area of compu- tational linguistics research. Early experiments (e.g., the 1954 Georgetown Exper- iment) successfully translated small numbers of sentences from one language to another (e.g. Russian to English). Experimenters then thought that MT would be solved by roughly the 1960s. Fast-forward to today, where the following is a free, online translation 1 of the above into French (minus parentheticals): 2 Les systmes de traduction automatique traduisent automatiquement dun de langage naturel dans des autres. Depuis les annees 50, ca ete un domaine actif de la recherche en mati`ere dinformatique linguistique. Les experiences tot ont avec succ`es traduit un nombre restreint de phrases dun langage `a lautre. Les experimentateurs ont alors pense que la TA serait resolue par rudement les annees 60. If you speak some French, youll see that this isnt all that bad. To get the flavor of what it is saying, heres the back-translation into English: The systems of automatic translation automatically translate one of natural language in others. Since the Fifties, it was an active field of search on compu- tational linguistics. The experiments early successfully translated a restricted number of sentences from one language to another. The experimenters then thought that MT would be solved by harshly the Sixties. Although this kind of translation isnt horrible, MT seems to be far from a solved problem. But even carrying out crummy translations (as they say) is very com- plicated. See whether you can spot (and indicate by labelling) some of the follow- ing linguistic phenomena in the little example from above: (Cross-lingual) Ambiguity A word in one language has more than one translation (possibly indicating more than one meaning) in another. Word-order Differences Words and phrases are in a different order in one lan- guage than in the other....
View Full Document

This note was uploaded on 04/21/2010 for the course PHILOSOPHY 150 taught by Professor Kerr during the Winter '10 term at Ohio State.

Page1 / 4

MTInClassHandout - L INGUISTICS 384: LANGUAGE AND COMPUTERS...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online