Name
Abstract | ||
|---|---|---|
New variations on the application of the passive-aggressive algorithm to statistical machine translation are developed and compared to previously existing approaches. In online adaptation, the system needs to adapt to real-world changing scenarios, where training and tuning only take place when the system is set-up for the first time. Post-edit information, as described by a given quality measure, is used as valuable feedback within the passive-aggressive framework, adapting the statistical models on-line. First, by modifying the translation model parameters, and alternatively, by adapting the scaling factors present in state-of-the-art SMT systems. Experimental results show improvements in translation quality by allowing the system to learn on a sentence-by-sentence basis. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1007/978-3-642-21257-4_30 | Pattern Recognition and Image Analysis |
Keywords | Field | DocType |
passive-aggressive algorithm,translation quality,translation model parameter,statistical machine translation,state-of-the-art smt system,passive-aggressive framework,statistical model,post-edit information,quality measure,on-line learning | Computer science,Machine translation,Statistical model,Artificial intelligence,Scaling,Online adaptation,Machine learning | Conference |
Volume | ISSN | Citations |
6669 | 0302-9743 | 1 |
PageRank | References | Authors |
0.36 | 11 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Pascual Martínez-Gómez | 1 | 61 | 7.36 |
| Germán Sanchis-Trilles | 2 | 101 | 16.95 |
| francisco casacuberta | 3 | 1439 | 161.33 |