Dependency Language Models for Transition-based Dependency Parsing

TL;DR

This paper introduces dependency language models to enhance transition-based dependency parsing, achieving state-of-the-art results on Chinese and competitive accuracy on English by integrating model-based features.

Contribution

The paper proposes a novel method of incorporating dependency language models into transition-based parsers to improve accuracy, demonstrating significant performance gains.

Findings

State-of-the-art accuracy on Chinese data

Improved UAS by 1 point on Chinese

Enhanced English parser accuracy by 0.5 points

Abstract

In this paper, we present an approach to improve the accuracy of a strong transition-based dependency parser by exploiting dependency language models that are extracted from a large parsed corpus. We integrated a small number of features based on the dependency language models into the parser. To demonstrate the effectiveness of the proposed approach, we evaluate our parser on standard English and Chinese data where the base parser could achieve competitive accuracy scores. Our enhanced parser achieved state-of-the-art accuracy on Chinese data and competitive results on English data. We gained a large absolute improvement of one point (UAS) on Chinese and 0.5 points for English.