Path Based Hierarchical Clustering on Knowledge Graphs

TL;DR

This paper introduces a new method for creating hierarchical clusters of subjects within knowledge graphs, using a tag hierarchy to improve the organization and reasoning capabilities of relational data.

Contribution

It presents a novel approach that constructs a tag hierarchy and assigns subjects to clusters, advancing taxonomy induction techniques for knowledge graphs.

Findings

Successfully induces coherent cluster hierarchies on real-world datasets

Demonstrates improved organization of relational data in knowledge graphs

Builds upon and extends previous taxonomy induction methods

Abstract

Knowledge graphs have emerged as a widely adopted medium for storing relational data, making methods for automatically reasoning with them highly desirable. In this paper, we present a novel approach for inducing a hierarchy of subject clusters, building upon our earlier work done in taxonomy induction. Our method first constructs a tag hierarchy before assigning subjects to clusters on this hierarchy. We quantitatively demonstrate our method's ability to induce a coherent cluster hierarchy on three real-world datasets.