Statistical Machine Translation (SMT) deals with automati cally mapping sentences in one human language (for example French) into another huma n language (such as English). The first language is called the source and the second language is called the target. This process can be thought of as a stochastic process. Ther e are many SMT variants, depending upon how translation is modelled. S ome approaches are in terms of a string-to-string mapping, some use trees-to-s trings, and some use treeto-tree models. All share in common the central idea that tra nslation is automatic, with models estimated from parallel corpora (source-targe t pairs) and also from monolingual corpora (examples of target sentences).
[1]
Robert L. Mercer,et al.
The Mathematics of Statistical Machine Translation: Parameter Estimation
,
1993,
CL.
[2]
Hermann Ney,et al.
Discriminative Training and Maximum Entropy Models for Statistical Machine Translation
,
2002,
ACL.
[3]
Salim Roukos,et al.
Bleu: a Method for Automatic Evaluation of Machine Translation
,
2002,
ACL.
[4]
Daniel Marcu,et al.
Statistical Phrase-Based Translation
,
2003,
NAACL.
[5]
David Chiang,et al.
A Hierarchical Phrase-Based Model for Statistical Machine Translation
,
2005,
ACL.