Monthly Archives: April 2012

Transducens group system for cross-lingual entailment, the second best performing in SemEval 2012

The system presented by the researchers of Transducens group Miquel Esplà-Gomis, Felipe Sánchez-Martínez, and Mikel L. Forcada, was the second best system in the cross-lingual textual entailiment (CLTE) for English-Spanish task according to the ranking published for the task 8 of SemEval 2012. CLTE is a natural language processing open problem which consists of detecting semantic entailment relations between texts written in different languages. The system presented by the Transducens’ researches uses a multiclass classifier informed by features obtained from machine translation.

Eleven grants from Google for the Apertium project

The free/open-source machine translation platform Apertium, originally created by the Transducens group, has been selected again by the Google Summer of Code initiative. This summer, eleven students from several countries around the world (out of the 1,200 total students participating in the programme) will get a grant of 5,000 USD from Google to work for Apertium for three months. In addition, two members of Transducens (Francis Tyers and Mikel L. Forcada) will participate in this edition of Google Summer of Code as mentors of some of these students.