DMoG : A Data-Based Morphological Guesser
|Článek ve sborníku
|Recent Advances in Slavonic Natural Language Processing (RASLAN 2021)
|Fakulta / Pracoviště MU
|Lemmatization; Morphological guesser; Morphological analysis; Morphological guessing
|We present a novel corpus-based approach to lemmatization of unknown words. The tool learns a?ix patterns from annotated data, and based on these patterns, it predicts other word forms that should be present in the corpus. A lemma candidate then comes from the pattern whose predictions are really found in the corpus. We present a prototype implementation and an initial evaluation on Czech, which shows promising results.