Large-scale Taxonomy Induction Using Entity and Word Embeddings

TL;DR

This paper introduces TIEmb, an unsupervised method that leverages entity and text embeddings to automatically induce large-scale taxonomies from web-based knowledge bases, reducing manual effort.

Contribution

It presents a novel approach for automatic taxonomy induction using embeddings, applied to large web-derived knowledge bases for specific domains.

Findings

Successfully extracted class hierarchies in Person and Place domains.

Demonstrated effectiveness of embeddings for taxonomy induction.

Reduced manual effort in building large-scale taxonomies.

Abstract

Taxonomies are an important ingredient of knowledge organization, and serve as a backbone for more sophisticated knowledge representations in intelligent systems, such as formal ontologies. However, building taxonomies manually is a costly endeavor, and hence, automatic methods for taxonomy induction are a good alternative to build large-scale taxonomies. In this paper, we propose TIEmb, an approach for automatic unsupervised class subsumption axiom extraction from knowledge bases using entity and text embeddings. We apply the approach on the WebIsA database, a database of subsumption relations extracted from the large portion of the World Wide Web, to extract class hierarchies in the Person and Place domain.