Pseudo-Projective Dependency Parsing

By Joakim Nivre, Jens Nilsson. Annual Meeting of the Association for Computational Linguistics, 2005

371 citations recorded by the source corpus.

In order to realize the full potential of dependency-based syntactic parsing, it is desirable to allow non-projective dependency structures. We show how a data-driven deterministic dependency parser, in itself restricted to projective structures, can be combined with graph transformation techniques to produce non-projective structures. Experiments using data from the Prague Dependency Treebank show that the combined system can handle non-projective constructions with a precision sufficient to yield a significant improvement in overall parsing accuracy. This leads to the best reported performance for robust non-projective parsing of Czech.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/b92c0e898b1fee243864176e18a2b50105be3e54

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Dependency-based Semantic Role Labeling of PropBank (2008, methodology): “To simplify implementation, we instead opted for the pseudo-projective approach (Nivre and Nilsson, 2005), in which nonprojective links are lifted up-wards in the tree to achieve projectivity, and special trace labels are used to enable recovery of the nonprojective links at parse time.”
A Latent Variable Model of Synchronous Syntactic-Semantic Parsing for Multiple Languages (2009, methodology): “We adopt the HEAD method of Nivre and Nilsson (2005) to de-projectivise syntactic dependencies outside of parsing.”
Algorithms for Deterministic Incremental Dependency Parsing (2008, methodology): “This corresponds to the Baseline condition in Nivre and Nilsson (2005).”
Characterizing the Errors of Data-Driven Dependency Parsing Models (2007, background): “Nivre and Nilsson (2005) showed how the restriction to projective dependency graphs could be lifted by using graph transformation techniques to pre-process training data and post-process parser output, so-called pseudo-projective parsing .”
CoNLL-X Shared Task on Multilingual Dependency Parsing (2006, methodology): “• The pseudo-projective approach (Nivre and Nilsson, 2005): Transform non-projective training trees to projective ones but encode the information necessary to make the inverse transformation in the DEPREL, so that this inverse transformation can also be carried out on the test trees (Nivre et al.,…”
Many Languages, One Parser (2016, methodology): “The parser uses the arc-standard transition system (Nivre, 2004).11 At each timestept, a transition ac- 11In a preprocessing step, we transform non-projective trees in the training treebanks to pseudo-projective trees usingthe "baseline" scheme in (Nivre and Nilsson, 2005).”
Non-Projective Dependency Parsing using Spanning Tree Algorithms (2005, background): “Nivre and Nilsson (2005) presented a parsing model that allows for the introduction of non-projective edges into dependency trees through learned edge transformations within their memory-based parser.”
Design and realization of a modular architecture for textual entailment (2013, methodology): “Another pipeline that is available for English and German is based on TreeTagger (Schmid 1994) and MaltParser (Nivre and Nilsson 2005).”
Online Learning of Approximate Dependency Parsing Algorithms (2006, methodology): “We can easily motivate this approximation by observing that even in non-projective languages like Czech and Danish, most trees are primarily projective with just a few non-projective edges (Nivre and Nilsson, 2005).”
Syntactic Annotations for the Google Books NGram Corpus (2012, methodology): “For treebanks with non-projective trees we use the pseudo-projective parsing technique to transform the treebank into projective structures (Nivre and Nilsson, 2005).”
Universal Stanford dependencies: A cross-linguistic typology (2014, background): “This kind of transform-detransform architecture is at present less common in dependency parsing, although Nilsson, Nivre & Hall (2006; 2007) do this for coordination and verb groups, and pseudo-projective parsing (Nivre and Nilsson, 2005) can also be seen as an instance of this architecture.”