Improving UD processing via satellite resources for morphology
Kaja Dobrovoljc, Tomaž Erjavec, Nikola Ljubešić · 2019
This paper presents the conversion of the reference language resources for Croatian and Slovenian morphology processing to UD morphological specifications.We show that the newly available training corpora and inflectional dictionaries improve the baseline stanfordnlp performance obtained on officially released UD datasets for lemmatization, morphology prediction and dependency parsing, illustrating the potential value of such satellite UD resources for languages with rich morphology.