2022/03/01 by Michael Rawson, Rawson, Michael, Samuel Dooley +6 · 1 citation
Computer Science · Mathematics · #Algebraic Topology (math.AT) #Algorithm #Artificial intelligence #Barcode #Cluster analysis #Computation and Language (cs.CL) #Computer science #Conceptualization #FOS: Computer and information sciences #FOS: Mathematics #Homotopy and Cohomology in Algebraic Topology #Mathematics #Natural language processing #Persistent homology #Simple (philosophy) #Theoretical computer science #Topological and Geometric Data Analysis #Topological data analysis #Topology (electrical circuits) #Word (group theory) #Word-sense disambiguation #cs.CL #math.AT
paper · pdf · doi:10.48550/arxiv.2203.00565
published in arXiv (Cornell University) (Cornell University)
arxiv created 2022/03/01 · openalex publication_date 2022/03/01 · arxiv updated 2022/03/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04
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.