Contextualization strategies for the qualitative-quantitative analysis of very large corpora

What is it about?

In recent years, both the availability of electronic resources (such as the BNC and the DWDS) and the possibility of compiling highly specialised large corpora (e.g. on the same topic, interaction type or communicative event) together with the development of ever more sophisticated corpus-linguistic tools, have opened new horizons in the study of language use and, consequently, in lexicography, translation and the teaching and learning of second/foreign languages. However, working with huge amounts of data and/or information remains a challenging issue, and search strategies are needed to cope more effectively with linguistic phenomena such as intralinguistic and interlinguistic variation in language use and polyfunctionality, which appear to be the rule rather than the exception in natural languages, not only in everyday language but in LSP communication, too. This paper will focus on different contextualization strategies. The examples offered, drawn from highly specialized subcorpora on the one hand and from very large general language corpora on the other, will illustrate once more the impact of variables such as domain, topic, text type etc., which can be observed even in the case of core vocabulary items. Furthermore it will discuss the potential offered by concgrams, which allow us to cope with many of the difficulties described above.

Featured Image

Why is it important?

Contextualization with the help of concgrams (e.g. triples of non-adjacent words) allows us to develop very effective powerful data-selection strategies.

Perspectives