Establishing sentential structure via realignments from small parallel corpora

Hits: 4927
Research areas: Year: 2015
Type of Publication: In Proceedings
  • 34, Vassiliki Pouli
Book title: Proceedings of the Fourth Workshop on Hybrid Approaches to Translation (HyTra)
Pages: 21-29
Address: Beijing, China
Organization: HYTRA-2015 Workshop [ held within ACL/IJCNLP 2015] Month: July 31
ISBN: 978-1-941643-67-9
The present article reports on efforts to improve the translation accuracy of a corpus-based hybrid MT system developed using the PRESEMT methodology. This methodology operates on a phrasal basis, where phrases are linguistically-motivated but are automatically determined via a dedicated module. Here, emphasis is placed on improving the structure of each translated sentence, by replacing the Example-Based MT approach originally used in PRESEMT with a sub-sentential approach. Results indicate that an improved accuracy can be achieved, as measured by objective metrics.