Semi-Supervised Feature Transformation for Dependency Parsing

In current dependency parsing models, conventional features (i.e. base features) defined over surface words and part-of-speech tags in a relatively high-dimensional feature space may suffer from the data sparseness problem and thus exhibit less discriminative power on unseen data. In this paper, we propose a novel semi-supervised approach to addressing the problem by transforming the base features into high-level features (i.e. meta features) with the help of a large amount of automatically parsed data. The meta features are used together with base features in our final parser. Our studies indicate that our proposed approach is very effective in processing unseen data and features. Experiments on Chinese and English data sets show that the final parser achieves the best-reported accuracy on the Chinese data and comparable accuracy with the best known parsers on the English data.

• Publication year

  2013

• Venue

  Conference on Empirical Methods in Natural Language Processing

• Publication date

  2013-10-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.18653/v1/d13-1129
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Utilizing Dependency Language Models for Graph-based Dependency Parsing Models

  2012cited by this paper
• A Separately Passive-Aggressive Training Algorithm for Joint POS Tagging and Dependency Parsing

  2012cited by this paper
• Joint Models for Chinese POS Tagging and Dependency Parsing

  2011cited by this paper
• Exploiting Web-Derived Selectional Preference to Improve Statistical Dependency Parsing

  2011cited by this paper
• Incremental Joint POS Tagging and Dependency Parsing in Chinese

  2011influential reference
• Transition-based Dependency Parsing with Rich Non-local Features

  2011cited by this paper
• Learning Condensed Feature Representations from Large Unsupervised Data Sets for Supervised Learning

  2011influential reference
• Efficient Third-Order Dependency Parsers

  2010cited by this paper
• Top Accuracy and Fast Dependency Parsing is not a Contradiction

  2010cited by this paper
• An Error-Driven Word-Character Hybrid Model for Joint Chinese Word Segmentation and POS Tagging

  2009cited by this paper
• An Empirical Study of Semi-supervised Structured Conditional Models for Dependency Parsing

  2009cited by this paper
• Improving Dependency Parsing with Subtrees from Auto-Parsed Data

  2009cited by this paper
• Semi-Supervised Sequential Labeling and Segmentation Using Giga-Word Scale Unlabeled Data

  2008cited by this paper
• Simple Semi-supervised Dependency Parsing

  2008influential reference
• A Tale of Two Parsers: Investigating and Combining Graph-based and Transition-based Dependency Parsing

  2008cited by this paper
• Characterizing the Errors of Data-Driven Dependency Parsing Models

  2007cited by this paper
• The CoNLL 2007 Shared Task on Dependency Parsing

  2007cited by this paper
• Dependency Parsing and Domain Adaptation with LR Models and Parser Ensembles

  2007cited by this paper
• Probabilistic Models for Action-Based Chinese Dependency Parsing

  2007cited by this paper
• Experiments with a Higher-Order Projective Dependency Parser

  2007influential reference
• Online Learning of Approximate Dependency Parsing Algorithms

  2006cited by this paper
• CoNLL-X Shared Task on Multilingual Dependency Parsing

  2006cited by this paper
• Reranking and Self-Training for Parser Adaptation

  2006cited by this paper
• A High-Performance Semi-Supervised Learning Method for Text Chunking

  2005influential reference
• Online Large-Margin Training of Dependency Parsers

  2005influential reference
• Deterministic Dependency Parsing of English Text

  2004cited by this paper
• Ultraconservative Online Algorithms for Multiclass Problems

  2003cited by this paper
• Statistical Dependency Analysis with Support Vector Machines

  2003cited by this paper
• Three New Probabilistic Models for Dependency Parsing: An Exploration

  1996cited by this paper
• A Maximum Entropy Model for Part-Of-Speech Tagging

  1996cited by this paper
• Building a Large Annotated Corpus of English: The Penn Treebank

  1993cited by this paper

• Semi-supervised Domain Adaptation for Dependency Parsing with Dynamic Matching Network

  2022cites this paper
• Semi-Supervised Methods for Out-of-Domain Dependency Parsing

  2018influential citation
• Word-Context Character Embeddings for Chinese Word Segmentation

  2017cites this paper
• Exploiting Vocabulary, Morphological, and Subtree Knowledge to Improve Chinese Syntactic Analysis

  2016cites this paper
• Improving Unsupervised Dependency Parsing with Knowledge from Query Logs

  2016cites this paper
• Dependency Language Models for Transition-based Dependency Parsing

  2016influential citation
• Exploiting meta features for dependency parsing and part-of-speech tagging

  2016cites this paper
• Structured Training for Neural Network Transition-Based Parsing

  2015cites this paper
• Domain Adaptation for Dependency Parsing via Self-Training

  2015cites this paper
• Web-scale Surface and Syntactic n-gram Features for Dependency Parsing

  2015cites this paper
• Distributed Feature Representations for Dependency Parsing

  2015influential citation
• Experiments on Semi-supervised Dependency Parsing of a Morphologically Rich Language

  2015influential citation
• Training with Auto-parsed Whole Trees

  2015cites this paper
• Training with Lexical Information

  2015cites this paper
• Overview of Semi-supervised Dependency Parsing Approaches

  2015cites this paper
• Evaluating Parsers with Dependency Constraints

  2015cites this paper
• Dependency Parse Reranking with Rich Subtree Features

  2014cites this paper
• Feature Embedding for Dependency Parsing

  2014cites this paper
• Fused Feature Representation Discovery for High-Dimensional and Sparse Data

  2014cites this paper
• Dependency Parsing with Structure Knowledge

  2014cites this paper
• Ambiguity-aware Ensemble Training for Semi-supervised Dependency Parsing

  2014influential citation