A Framework for Understanding the Role of Morphology in Universal Dependency Parsing

Mathieu Dehouck (1, 2) , Pascal Denis (2, 1)

1 MAGNET - Machine Learning in Information Networks
2 CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189

This paper presents a simple framework for characterizing morphological complexity and how it encodes syntactic information. In particular, we propose a new measure of morphosyntactic complexity in terms of governordependent preferential attachment that explains parsing performance. Through experiments on dependency parsing with data from Universal Dependencies (UD), we show that representations derived from morphological attributes deliver important parsing performance improvements over standard word form embeddings when trained on the same datasets. We also show that the new morphosyntactic complexity measure is predictive of the gains provided by using morphological attributes over plain forms on parsing scores, making it a tool to distinguish languages using morphology as a syntactic marker from others.

Domaines: Informatique et langage [cs.CL]; Apprentissage [cs.LG]; Traitement du texte et du document; Linguistique

Mathieu Dehouck, Pascal Denis. A Framework for Understanding the Role of Morphology in Universal Dependency Parsing. EMNLP 2018 - Conference on Empirical Methods in Natural Language Processing, Oct 2018, Brussels, Belgium. ⟨hal-01943934⟩