Rough Sets for Explainability of Spectral Graph Clustering

TL;DR

This paper enhances spectral graph clustering explainability by integrating rough set theory, addressing challenges posed by spectral embedding complexity, ambiguous content, and stochastic variability in clustering results.

Contribution

It introduces a novel explanation method based on rough set theory to improve interpretability of spectral graph clustering outputs.

Findings

Improved explainability of spectral clustering results.

Effective handling of ambiguous and contentless documents.

Enhanced robustness against stochastic variations.

Abstract

Graph Spectral Clustering methods (GSC) allow representing clusters of diverse shapes, densities, etc. However, the results of such algorithms, when applied e.g. to text documents, are hard to explain to the user, especially due to embedding in the spectral space which has no obvious relation to document contents. Furthermore, the presence of documents without clear content meaning and the stochastic nature of the clustering algorithms deteriorate explainability. This paper proposes an enhancement to the explanation methodology, proposed in an earlier research of our team. It allows us to overcome the latter problems by taking inspiration from rough set theory.