Ensemble Clustering with Logic Rules

TL;DR

This paper introduces a novel ensemble clustering method that employs logic rule ensembles to partition data hierarchically, utilizing similarity matrices and validity measures to improve clustering quality and determine the optimal number of clusters.

Contribution

It adapts logic rule ensembles from supervised learning to unsupervised and semi-supervised clustering, providing a new approach for hierarchical data partitioning.

Findings

Effective clustering with logic rule ensembles

Improved cluster validity assessment

Hierarchical clustering performance demonstrated

Abstract

In this article, the logic rule ensembles approach to supervised learning is applied to the unsupervised or semi-supervised clustering. Logic rules which were obtained by combining simple conjunctive rules are used to partition the input space and an ensemble of these rules is used to define a similarity matrix. Similarity partitioning is used to partition the data in an hierarchical manner. We have used internal and external measures of cluster validity to evaluate the quality of clusterings or to identify the number of clusters.