Sequence Tagging for Verb Conjugation in Romanian
Liviu P. Dinu, Octavia-Maria Şulea, Vlad Niculae · Recent Advances in Natural Language Processing · 2013
Verbs in Romanian sometimes manifest local irregularities in the form of alternating letters. We present a sequence tagging based method for learning stem alternations and ending sequences. Supervised training is based on a morphological dictionary, with a few regular expression paradigms encoded by hand. Our best model improves upon previous machine learning approaches to Romanian verb conjugation, and can generalize to unseen paradigms that can be constructed as variations of the ones in the training set.