Changes between Version 2 and Version 3 of en/MachineTranslation
- Timestamp:
- Jun 5, 2014, 3:16:21 PM (10 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
en/MachineTranslation
v2 v3 6 6 * Statistical machine translation 7 7 * Extension of translation memories 8 * Domain-speci 9 c machine translation 8 * Domain-specic machine translation 10 9 * Machine translation between close languages 11 10 * Sub-word level machine translation … … 19 18 * baseline SMT available for everyone 20 19 * languages with high number of wordforms need special treatment 21 * '''d ělám, děláš, dělal, dělajícímu, dělaje, děláním, ...'''20 * '''dělám, děláš, dělal, dělajícímu, dělaje, děláním, ...''' 22 21 23 22 '''magas, magasabb, legmagasabb, legeslegmagasabb, ...''' … … 30 29 31 30 [[Image(/trac/research/raw-attachment/wiki/en/MachineTranslation/word_matrix1.png)]] 31 32 32 33 [[Image(/trac/research/raw-attachment/wiki/en/MachineTranslation/word_matrix2.png)]] 34 35 33 36 [[Image(/trac/research/raw-attachment/wiki/en/MachineTranslation/word_matrix3.png)]] 34 37 35 38 36 == Domain-speci 37 c machine translation == 39 == Domain-specic machine translation == 38 40 * straightforward way of increasing quality of MT 39 * domain-speci 40 c corpora can be downloaded on demand 41 * domain-specic corpora can be downloaded on demand 41 42 * separate models for each domain: ''' sports, cooking, gardening''' 42 43 * one sense per domain: '''bat''' … … 46 47 * translations of 47 48 * product details, product descriptions in e-shops, 48 * manuals, warranty certi 49 cates, 49 * manuals, warranty certicates, 50 50 * user interface localizations, ... 51 51 … … 54 54 * West and South Slavic languages: Czech, Slovak, Polish, Serbian, Croatian, Slovene 55 55 * MT mainly on word level, structure is very similar 56 * di 57 erences can be described systematically by rules: 58 '''hraje na klavíri''' <-> '''hraje na klavír''' 56 * dierences can be described systematically by rules: 57 '''hraje na klavíri''' <-> '''hraje na klavír''' 59 58 * billion-word corpora available for these languages 60 59 * dictionaries can be generated semi-automatically … … 79 78 == Conclusions == 80 79 * generating new segments for translation memories 81 * domain-speci 82 c translation 80 * domain-specic translation 83 81 * translation between close languages 84 82 * sub-word level translation