An Environment for Learner Corpus Research and Error Analysis: The Study of Determiner Errors in Basque

Methodology

As regards error analysis methodology, language errors are usually collected from learner corpora, tagged and classified. Then, analysing those texts and the tagged errors, it is possible to make some conclusions on the learners’ interlanguage (IL), the developing language of second language (2L) learners. Interlanguage often differs from the target language (L2), and the annotation of such corpora is an important means of accessing its unique characteristics (Granger, 2003a).

However, manual annotation is a difficult and costly job. That is why many computer tools for language learner error analysis have been developed (Dagneaux, Denness, & Granger, 1998; Meunier, 1998) which enable the error tagging to be performed more accurately, quickly and easily. Apart from annotating errors for language learning and teaching purposes, our idea when developing the environment here presented was to take the advantage of those annotated corpora to carry out the automatic error detection, diagnosis and correction of the tagged error examples.