1 |
Boosting Performance of Weak MT Engines Automatically: Using MT Output to Align Segments & Build Statistical Post-Editors
|
|
|
|
In: http://www.mt-archive.info/EAMT-2008-Voss.pdf
|
|
Abstract:
Abstract. This paper addresses the practical challenge of improving existing, op-erational translation systems with relatively weak, black-box MT engines when higher quality MT engines are not available and only a limited quantity of online re-sources is available. Recent research results show impressive performance gains in translating between Indo-European languages when chaining mature, existing rule-based MT engines and post-MT editors built automatically with limited amounts of parallel data. We show that this hybrid approach of serially composing or “chaining” an MT engine and automated post-MT editor---when applied to much weaker lexi-con-based and rule-based MT engines, translating across the more widely divergent languages of Urdu and English, and given limited amounts of document-parallel only training data---will yield statistically significant boosts in translation quality up to the 50K of parallel segments in training the post-editor, but not necessarily be-yond that.
|
|
URL: http://www.mt-archive.info/EAMT-2008-Voss.pdf http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.550.5597
|
|
BASE
|
|
Hide details
|
|
2 |
Arabic strings for analysis by BMA (see Fig. 1)
|
|
|
|
In: http://aclweb.org/anthology-new/W/W08/W08-0511.pdf
|
|
BASE
|
|
Show details
|
|
3 |
Exploitation of an Arabic Language Resource for MT Evaluation: Using Buckwalter-based Lookup Tool to Augment CMU Alignment Algorithm
|
|
|
|
In: http://www.mt-archive.info/LREC-2008-Voss.pdf
|
|
BASE
|
|
Show details
|
|
4 |
Exploitation of an Arabic Language Resource for MT Evaluation: Using Buckwalter-based Lookup Tool to Augment CMU Alignment Algorithm
|
|
|
|
In: http://www.lrec-conf.org/proceedings/lrec2008/pdf/887_paper.pdf
|
|
BASE
|
|
Show details
|
|
|
|