Total instances annotated in both training and test corpora.
The method uses a measure of the specificity of a terminology candidate with respect to the target domain via comparative analysis across different corpora.
My methods are predominantly quantitative, and my primary data comprise structured electronic corpora, which offer a solid basis for chronological comparisons.
The research reported in the second paper is motivated the task of automatic knowledge acquisition from large corpora.
Note, that this cautious measure may lead to underestimating the errors in the corpora.
Like all other research designs, using large corpora and concordance programs opens up specific potential but also has limitations.
Next, a set of textual corpora was compiled from which example sentences could be gathered.
However, the overall rates of orthography influence are low in all three corpora.
