Fast and extensible phrase scoring for Statistical Machine Translation
2010 (English)In: Prague Bulletin of Mathematical Linguistics, ISSN 0032-6585, E-ISSN 1804-0462, no 93, 87-96 p.Article in journal (Refereed) Published
Existing tools for generating phrase tables for phrase-based Statistical Machine Translation (SMT) are generally optimised towards low memory use to allow processing of large corpora with limited memory. Whilst being a reasonable design choice, this approach does not make optimal use of resources when the sufficient memory is available. We present memscore, a new open-source tool to score phrases in memory. Besides acting as a faster drop-in replacement for existing software, it implements a number of standard smoothing techniques and provides a platform for easy experimentation with new scoring methods.
Place, publisher, year, edition, pages
Prague: Charles University, Institute of Formal and Applied Linguistics , 2010. no 93, 87-96 p.
IdentifiersURN: urn:nbn:se:uu:diva-154487DOI: 10.2478v10108-010-0007-5OAI: oai:DiVA.org:uu-154487DiVA: diva2:420756
In Proceedings of MT Marathon 2010, Dublin (http://www.mtmarathon2010.info/web/Welcome.html).2011-06-072011-06-052013-09-18Bibliographically approved