Hybrid Models

0
ID: 
5.5
Task leader: 
cristina.españa
Assignees: 
cristina.españa
Assignees: 
lluis.marquez
Dependencies: 
Baseline systems
Dependencies: 
Machine Translation Systems
Dependencies: 
Patent Corpora
Dependencies: 
Patents Retrieval System
Status: 
Ongoing

Development of the hybrid models. Some ideas to be explored:

  • Hard integration. Force fixed GF translations within a SMT system.
  • Soft integration I. Led by SMT. GF partial output, as phrase pairs, is integrated as a discriminative probability feature model in a phrase-based SMT system.
  • Soft integration II. Led by SMT. GF partial output, as tree fragment pairs, is integrated as a discriminative probability model in a syntax-based SMT system.
  • Soft integration III. Led by GF. Complement with SMT options the GF translation structure and perform statistical search to find the final translation.

M24. UPC. (Corresponding to MS7 and D52, M24)

Comments

Including syntactic information from GF into SMT

As the reviewers suggested we could also include syntactic information from GF into SMT as additional features. A first thing to check could be the addition of PoS to the pure word translation.