Morphological and Syntactic Case in Statistical Dependency Parsing

Abstract

Morphologically rich languages that exhibit free word order frequently use case systems to mark the function of nominal elements, especially for the core argument functions of a verb. The standard pipeline approach in syntactic dependency parsing assumes a complete disambiguation of morphological (case) information prior to automatic syntactic analysis. Parsing experiments on Czech, German, and Hungarian show that this approach is susceptible to propagating morphological annotation errors when parsing languages that show syncretism in their morphological case paradigms. This article demonstrates how by interpreting (underspecified) case as a filtering device that restricts the options for syntactic analysis, carefully designed morphosyntactic constraints can restrict the search space of a statistical dependency parser, and prevent solutions that would violate the syntactic restrictions overtly marked in the morphology of the words in a given sentence.