Automatic Annotation of Morpho-Syntactic Dependencies in a Modern Hebrew Treebank

Abstract

"Morpho-syntactic dependencies between sentence constituents are an inseparable part of syntactic analysis, in particular in Semitic languages. In those languages, because of the relatively free order of certain constituents, morpho-syntactic agreement features are sometimes the main clue for computational parsing models (Tsarfaty and Sima’an, 2007; Tsarfaty and Sima’an, 2008). However, despite their centrality for syntactic analysis, morpho-syntactic dependencies have so far not been annotated in Hebrew resources. In particular, such dependencies were not annotated in early versions of the Modern Hebrew Treebank (MHT, Sima’an et al. (2001)), which to date is the only publicly available resource with syntactic analyses for Modern Hebrew. By developing a method for automatically adding dependency annotations to a Modern Hebrew treebank, the MHT project has aimed to contribute to treebank development for Semitic languages as well as for other languages"