Query-Based Named Entity Recognition

TL;DR

This paper introduces a query-based approach to named entity recognition that effectively handles overlapping entities and achieves state-of-the-art results across multiple datasets in English and Chinese.

Contribution

The paper proposes a novel query-based NER method that addresses overlapping entities and improves performance by leveraging question-answering techniques.

Findings

Sets new SOTA on five NER datasets

Effectively handles overlapping and nested entities

Improves extraction accuracy with query encoding

Abstract

In this paper, we propose a new strategy for the task of named entity recognition (NER). We cast the task as a query-based machine reading comprehension task: e.g., the task of extracting entities with PER is formalized as answering the question of "which person is mentioned in the text ?". Such a strategy comes with the advantage that it solves the long-standing issue of handling overlapping or nested entities (the same token that participates in more than one entity categories) with sequence-labeling techniques for NER. Additionally, since the query encodes informative prior knowledge, this strategy facilitates the process of entity extraction, leading to better performances. We experiment the proposed model on five widely used NER datasets on English and Chinese, including MSRA, Resume, OntoNotes, ACE04 and ACE05. The proposed model sets new SOTA results on all of these datasets.