The Role of Global and Local Context in Named Entity Recognition

TL;DR

This paper investigates how incorporating global document context, beyond local context, improves Named Entity Recognition performance in long documents using transformer-based models.

Contribution

It demonstrates that retrieving global document context significantly enhances NER accuracy, highlighting the need for better context retrieval methods.

Findings

Global context retrieval improves NER performance

Local context alone is insufficient for long documents

Further research needed on context retrieval techniques

Abstract

Pre-trained transformer-based models have recently shown great performance when applied to Named Entity Recognition (NER). As the complexity of their self-attention mechanism prevents them from processing long documents at once, these models are usually applied in a sequential fashion. Such an approach unfortunately only incorporates local context and prevents leveraging global document context in long documents such as novels, which might hinder performance. In this article, we explore the impact of global document context, and its relationships with local context. We find that correctly retrieving global document context has a greater impact on performance than only leveraging local context, prompting for further research on how to better retrieve that context.