Hierarchical Bracketing Encodings for Dependency Parsing as Tagging
By: Ana Ezquerro , David Vilares , Anssi Yli-Jyrä and more
Potential Business Impact:
Makes computer language understanding more efficient.
We present a family of encodings for sequence labeling dependency parsing, based on the concept of hierarchical bracketing. We prove that the existing 4-bit projective encoding belongs to this family, but it is suboptimal in the number of labels used to encode a tree. We derive an optimal hierarchical bracketing, which minimizes the number of symbols used and encodes projective trees using only 12 distinct labels (vs. 16 for the 4-bit encoding). We also extend optimal hierarchical bracketing to support arbitrary non-projectivity in a more compact way than previous encodings. Our new encodings yield competitive accuracy on a diverse set of treebanks.
Similar Papers
A Systematic Comparison of Syntactic Representations of Dependency Parsing
Computation and Language
Makes computer language understanding better for some languages.
Hierarchical Bayesian Operator-induced Symbolic Regression Trees for Structural Learning of Scientific Expressions
Methodology
Finds science rules from messy data.
Dependency Parsing with the Structuralized Prompt Template
Computation and Language
Helps computers understand sentence structure better.