Topological Data Analysis for Word Sense Disambiguation

TL;DR

This paper introduces a novel unsupervised method for word sense disambiguation using topological data analysis, specifically persistent homology, demonstrating promising results on the SemCor dataset.

Contribution

It applies advanced topological concepts to NLP, offering a new perspective beyond traditional clustering methods for word sense induction.

Findings

Low relative error on word sense induction tasks

Demonstrates the effectiveness of topological algorithms in NLP

Encourages further research in topological data analysis for language processing

Abstract

We develop and test a novel unsupervised algorithm for word sense induction and disambiguation which uses topological data analysis. Typical approaches to the problem involve clustering, based on simple low level features of distance in word embeddings. Our approach relies on advanced mathematical concepts in the field of topology which provides a richer conceptualization of clusters for the word sense induction tasks. We use a persistent homology barcode algorithm on the SemCor dataset and demonstrate that our approach gives low relative error on word sense induction. This shows the promise of topological algorithms for natural language processing and we advocate for future work in this promising area.