70  Last view: 2023-05-09

Morphological analyser, lemmatiser and synthesiser for Lithuanian

Lemuoklis is a morphological analyzer, lemmatizer and tagger for Lithuanian. It assigns its lemma (or several hypothetical lemmas) to each token in a text and performs its morphological analysis. A word form is characterized grammatically by a combination of properties with respect to 13 categories: part of speech, aspect, reflexiveness, voice, mood, tense, group, degree, definiteness, gender, number, case and person. The database of lexical and grammatical information of the program consists of six lexicons. In Lemuoklis, morphological rules are expressed in the form of digital tables. The tables represent
graph structures that define both collections of affixes and collections of grammatical properties. Using morphological rules together with word– root lexicons enables us to analyse milliards of theoretically available Lithuanian written forms.