Semi-supervised Dependency Parsing using Lexical Affinities

Seyed Abolghasem Mirroshandel (1) , Alexis Nasr (1) , Joseph Le Roux (2)

1 LIF - Laboratoire d'informatique Fondamentale de Marseille - UMR 6166
2 LIPN - Laboratoire d'Informatique de Paris-Nord

Treebanks are not large enough to reliably model precise lexical phenomena. This deficiency provokes attachment errors in the parsers trained on such data. We propose in this paper to compute lexical affinities, on large corpora, for specific lexico-syntactic configurations that are hard to disambiguate and introduce the new information in a parser. Experiments on the French Treebank showed a relative decrease of the error rate of 7.1% Labeled Accuracy Score yielding the best pars- ing results on this treebank.

Domaines: Informatique et langage [cs.CL]

Seyed Abolghasem Mirroshandel, Alexis Nasr, Joseph Le Roux. Semi-supervised Dependency Parsing using Lexical Affinities. ACL, 2012, South Korea. ⟨hal-00702486⟩