Discriminative Phrase-Based Statistical Machine Transation

  • Osborne, Miles (Principal Investigator)

Project Details

Key findings

We developed the first large-scale discriminative translation systems, introducing to the field the ideas of maximum-translation and maximum-derivation decoding and training, Additionally we showed how non-parametric methods could be used to induce translation
models from parallel data.
StatusFinished
Effective start/end date1/02/0731/01/10

Funding

  • EPSRC: £336,578.00

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.