Algorithms for Deterministic Incremental Dependency Parsing

6.1 Experimental Setup

1. How did Nivre (2008) evaluate the four parsing algorithms?
   On the 13 CoNLL-X shared task languages, with MaltParser and SVM classifiers (quadratic kernel) trained on oracle transition sequences from gold trees, using the same simple feature model for all. Seven parsers per language: non-projective list-based, and the three projective algorithms in strictly projective and pseudo-projective versions. Metrics: LAS, UAS, label accuracy.
   • Nivre, 2008, Algorithms for Deterministic Incremental Dependency Parsing (Computational Linguistics), §6.1, pp. 537–541
2. What is pseudo-projective parsing?
   Train a projective parser on projectivized data: non-projective arcs are lifted to the nearest permissible ancestor of the real head, with the real head’s label encoded in the arc label. After parsing, arcs with augmented labels are lowered by searching the head’s descendants for a node with that label (Nivre and Nilsson 2005).
   • Nivre, 2008, Algorithms for Deterministic Incremental Dependency Parsing (Computational Linguistics), §6.1, pp. 537–541
3. How does a classifier-based parser handle a predicted transition whose preconditions fail?
   It performs a default transition with no preconditions instead: SHIFT for the stack-based algorithms, NO-ARC for the list-based ones. Classifiers rarely predict illegal transitions.
   • Nivre, 2008, Algorithms for Deterministic Incremental Dependency Parsing (Computational Linguistics), §6.1, pp. 537–541