Design and development of Iberia: a corpus of scientific Spanish

Abstract

Iberia is a synchronic corpus of scientific Spanish designed mainly for terminological studies. In this paper, we describe its design and the infrastructure for its acquisition, processing and exploitation, including mark-up, linguistic annotation, indexing and the user interface. Two pre-processing tasks affecting a large number of words are described in detail: de-hyphenation and identification of text fragments in other languages. We also show how some of the reported statistics, namely, dispersion and association, are used for research on lexis.