Croatian Morphological Lexicon v4.6
The Croatian Morphological Lexicon is an inflectional lexicon generated automatically by Croatian Inflectional Generator from ca 110,000 lemmas yielding over 4,000,000 word forms. It has been a result of the group lead by Prof. Marko Tadić on the basis of theoretical background published in 1992 (see Tadić 1994 above). The initial set of lemmas was collected from several existing Croatian mono- and bi-lingual dictionaries, while additional entries were collected via corpus or by means of automatic enlargement of the initial list of lemmas (see Bekavac, Šojat 2005, and Oliver, Tadić 2004 above). The automatically generated output was corrected for known systemic errors, encoded in utf-8 and stored in MulTextEast Lexica format: lemma[TAB]word-form[TAB]MSD. The MSD-tagset is conformant with the MulTextEast v4.0 reccomendations for Croatian language. However, some additions exist: in surnames gender is left unspecified (-), additional subclassification of adverbials has been introduced etc.
At the moment the Croatian Morphological Lexicon is a pseudolexicon, accessible only through the Croatian Lemmatisation Server web query interface or php script call.
- Headword lists from different Croatian monolingual dictionaries, Croatian National Corpus, Croatian Web-Corpus
- Croatian Word-Forms Generator