Some Aspects of the Morphological Processing of Bulgarian

This paper demonstrates the modelling of morphological knowledge in Bulgarian and applications of the created data sets in an integrated framework for production and manipulation of language resources. The production scenario is exemplified by the Bulgarian verb as the morphologically richest and most problematic part-of-speech category. The definition of the set of morphosyntactic specifications for verbs in the lexicon is described. The application of the tagset in the automatic morphological analysis of text corpora is accounted for. A Type Model of Bulgarian verbs handling the attachment of short pronominal elements to verbs, is presented.