Dependency Parsing as Head Selection

2016/06/03 by Xingxing Zhang, Zhang, Xingxing, Jianpeng Cheng +3 · 1 citation
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Text Readability and Simplification #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1606.01280

openalex publication_date 2016/06/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Conventional graph-based dependency parsers guarantee a tree structure both during training and inference. Instead, we formalize dependency parsing as the problem of independently selecting the head of each word in a sentence. Our model which we call DeNSe (as shorthand for \bf Dependency \bf Neural \bf Selection) produces a distribution over possible heads for each word using features obtained from a bidirectional recurrent neural network. Without enforcing structural constraints during training, DeNSe generates (at inference time) trees for the overwhelming majority of sentences, while non-tree outputs can be adjusted with a maximum spanning tree algorithm. We evaluate DeNSe on four languages (English, Chinese, Czech, and German) with varying degrees of non-projectivity. Despite the simplicity of the approach, our parsers are on par with the state of the art.

Citations

• Adam: A Method for Stochastic Optimization
• Recurrent Neural Network Regularization
• Multi-task Sequence to Sequence Learning
• Recurrent Dropout Without Memory Loss
• Structured Training for Neural Network Transition-Based Parsing

Cited by

• Non-Projective Dependency Parsing via Latent Heads Representation (LHR)

Related

• Non-Projective Dependency Parsing via Latent Heads Representation (LHR)
• A Fundamental Algorithm for Dependency Parsing (With Corrections)
• A Re-ranking Model for Dependency Parser with Recursive Convolutional Neural Network
• Parsing as Reduction
• Read, Tag, and Parse All at Once, or Fully-neural Dependency Parsing