Title | ||
---|---|---|
Better statistical estimation can benefit all phrases in phrase-based statistical machine translation |
Abstract | ||
---|---|---|
The heuristic estimates of conditional phrase translation probabilities are based on frequency counts in a word-aligned parallel corpus. Earlier attempts at more principled estimation using Expectation-Maximization (EM) under perform this heuristic. This paper shows that a recently introduced novel estimator based on smoothing might provide a good alternative. When all phrase pairs are estimated (no length cut-off), this estimator slightly outperforms the heuristic estimator. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/SLT.2008.4777884 | Goa |
Keywords | Field | DocType |
expectation-maximisation algorithm,language translation,smoothing methods,conditional phrase translation probabilities,expectation-maximization,phrase-based statistical machine translation,smoothing methods,statistical estimation,word-aligned parallel corpus,Parameter Estimation,Smoothing Methods,Transduction | Heuristic,Language translation,Pattern recognition,Computer science,Machine translation,Phrase,Speech recognition,Smoothing,Artificial intelligence,Estimation theory,Estimator | Conference |
ISBN | Citations | PageRank |
978-1-4244-3472-5 | 1 | 0.36 |
References | Authors | |
11 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Khalil Sima'an | 1 | 443 | 50.32 |
Markos Mylonakis | 2 | 24 | 2.45 |
Sima'an, K. | 3 | 1 | 0.36 |