Unified Lattice Graph Fusion for Chinese Named Entity Recognition

TL;DR

This paper introduces ULGF, a novel graph-based method that explicitly models semantic and boundary relations in Chinese NER by converting lattice structures into unified graphs, improving recognition accuracy.

Contribution

The paper proposes a unified lattice graph fusion approach that captures semantic and boundary relations in Chinese NER, integrating multiple graph-based attention and cross-gating layers.

Findings

Outperforms existing methods on four Chinese NER datasets

Effectively models semantic and boundary relations in lattice structures

Leverages lexicon entity classification as auxiliary task

Abstract

Integrating lexicon into character-level sequence has been proven effective to leverage word boundary and semantic information in Chinese named entity recognition (NER). However, prior approaches usually utilize feature weighting and position coupling to integrate word information, but ignore the semantic and contextual correspondence between the fine-grained semantic units in the character-word space. To solve this issue, we propose a Unified Lattice Graph Fusion (ULGF) approach for Chinese NER. ULGF can explicitly capture various semantic and boundary relations across different semantic units with the adjacency matrix by converting the lattice structure into a unified graph. We stack multiple graph-based intra-source self-attention and inter-source cross-gating fusion layers that iteratively carry out semantic interactions to learn node representations. To alleviate the over-reliance on word information, we further propose to leverage lexicon entity classification as an auxiliary task. Experiments on four Chinese NER benchmark datasets demonstrate the superiority of our ULGF approach.