Named Entity Recognition as Dependency Parsing

TL;DR

This paper introduces a novel dependency parsing approach for Named Entity Recognition that effectively handles both nested and flat entities, achieving state-of-the-art results across multiple datasets.

Contribution

It applies graph-based dependency parsing with a biaffine model to improve NER, especially for nested entities, demonstrating superior performance over existing methods.

Findings

Achieved state-of-the-art results on 8 NER corpora.

Improved accuracy by up to 2.2 percentage points.

Effectively models nested and flat entities.

Abstract

Named Entity Recognition (NER) is a fundamental task in Natural Language Processing, concerned with identifying spans of text expressing references to entities. NER research is often focused on flat entities only (flat NER), ignoring the fact that entity references can be nested, as in [Bank of [China]] (Finkel and Manning, 2009). In this paper, we use ideas from graph-based dependency parsing to provide our model a global view on the input via a biaffine model (Dozat and Manning, 2017). The biaffine model scores pairs of start and end tokens in a sentence which we use to explore all spans, so that the model is able to predict named entities accurately. We show that the model works well for both nested and flat NER through evaluation on 8 corpora and achieving SoTA performance on all of them, with accuracy gains of up to 2.2 percentage points.