Strong Heuristics for Named Entity Linking

TL;DR

This paper introduces simple, scalable heuristics for named entity linking in large news corpora, achieving high disambiguation accuracy and outperforming some state-of-the-art unsupervised methods.

Contribution

It proposes lightweight heuristics for entity disambiguation that are effective, scalable, and competitive with advanced unsupervised and zero-shot approaches.

Findings

Disambiguates 94% of mentions in Quotebank

Achieves 63% disambiguation on AIDA-CoNLL

Outperforms Eigenthemes and mGENRE in benchmarks

Abstract

Named entity linking (NEL) in news is a challenging endeavour due to the frequency of unseen and emerging entities, which necessitates the use of unsupervised or zero-shot methods. However, such methods tend to come with caveats, such as no integration of suitable knowledge bases (like Wikidata) for emerging entities, a lack of scalability, and poor interpretability. Here, we consider person disambiguation in Quotebank, a massive corpus of speaker-attributed quotations from the news, and investigate the suitability of intuitive, lightweight, and scalable heuristics for NEL in web-scale corpora. Our best performing heuristic disambiguates 94% and 63% of the mentions on Quotebank and the AIDA-CoNLL benchmark, respectively. Additionally, the proposed heuristics compare favourably to the state-of-the-art unsupervised and zero-shot methods, Eigenthemes and mGENRE, respectively, thereby serving as strong baselines for unsupervised and zero-shot entity linking.