Homonymy Information for English WordNet

TL;DR

This paper introduces a novel method for annotating homonymy in WordNet by linking it to the Oxford English Dictionary using transformer-based embeddings, achieving high accuracy and providing a valuable resource.

Contribution

It presents a new approach that leverages language models to automatically annotate homonymy in WordNet through dictionary alignment, improving over previous clustering methods.

Findings

Achieved an F1 score of 0.97 on evaluation data.

Produced a high-quality homonymy annotation layer for WordNet.

Demonstrated the effectiveness of embedding-based dictionary alignment.

Abstract

A widely acknowledged shortcoming of WordNet is that it lacks a distinction between word meanings which are systematically related (polysemy), and those which are coincidental (homonymy). Several previous works have attempted to fill this gap, by inferring this information using computational methods. We revisit this task, and exploit recent advances in language modelling to synthesise homonymy annotation for Princeton WordNet. Previous approaches treat the problem using clustering methods; by contrast, our method works by linking WordNet to the Oxford English Dictionary, which contains the information we need. To perform this alignment, we pair definitions based on their proximity in an embedding space produced by a Transformer model. Despite the simplicity of this approach, our best model attains an F1 of .97 on an evaluation set that we annotate. The outcome of our work is a high-quality homonymy annotation layer for Princeton WordNet, which we release.