Condorcet's Jury Theorem for Consensus Clustering
Brijnesh Johannes Jain · arXiv (Cornell University) · 2016
The goal of consensus clustering is to improve the quality of clustering by combining a sample of partitions of a dataset to a single consensus partition. This contribution extends Condorcet's Jury Theorem to the mean partition approach of consensus clustering. As a consequence of the proposed result, we challenge and reappraise the role of diversity in consensus clustering.