title

Extracting bilingual vocabulary

In the process of extracting terminology from a language corpus, parallel bilingual corpora are a field with their own peculiarities.

UZEI has developed this application in this respect:

Lex2: based on parallel corpora, this tool detects and automatically extracts pairs of terms in Basque and Spanish (see Lex2).

These automatically extracted pairs of terms can make up a bilingual vocabulary.
Translation memories are a clear example of a parallel corpus. Consequently, Lex2 may be of great use, for example, in the extraction of terminology used in these translation memories and for detecting and correcting possible inconsistencies between different correlations of the terms used.