Exploiting the corpus data in specification of language variants, esp. in orthography, word derivation and morphology.
