Abstract | ||
---|---|---|
This paper describes an efficient method to extract large n-best lists from a word graph produced by a statistical machine translation system. The extraction is based on the k shortest paths algorithm which is efficient even for very large k. We show that, although we can generate large amounts of distinct translation hypotheses, these numerous candidates are not able to significantly improve overall system performance. We conclude that large n-best lists would benefit from better discriminating models. |
Year | Venue | Keywords |
---|---|---|
2007 | HLT-NAACL (Short Papers) | statistical machine translation system,large k,numerous candidate,word graph,efficient method,large n-best list,large amount,k shortest paths algorithm,overall system performance,distinct translation hypothesis,shortest path algorithm |
Field | DocType | Citations |
Computer science,Machine translation system,Natural language processing,Artificial intelligence,Word graph,Machine learning | Conference | 5 |
PageRank | References | Authors |
0.44 | 7 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sasa Hasan | 1 | 245 | 17.35 |
Richard Zens | 2 | 2094 | 114.40 |
Hermann Ney | 3 | 14178 | 1506.93 |