A Distributional Approach for Soft Clustering Comparison and Evaluation

  title={A Distributional Approach for Soft Clustering Comparison and Evaluation},
  author={Andrea Campagner and Davide Ciucci and Thierry Denoeux},
. The development of external evaluation criteria for soft clustering (SC) has received limited attention: existing methods do not provide a general approach to extend comparison measures to SC, and are unable to account for the uncertainty represented in the results of SC algorithms. In this article, we propose a general method to address these limitations, grounding on a novel interpretation of SC as distributions over hard clusterings, which we call distributional measures . We provide an in… 

Tables from this paper



Objective Criteria for the Evaluation of Clustering Methods

This article proposes several criteria which isolate specific aspects of the performance of a method, such as its retrieval of inherent structure, its sensitivity to resampling and the stability of its results in the light of new data.

A new Mallows distance based metric for comparing clusterings

This paper proposes a measure for comparing clustering results to tackle two issues insufficiently addressed or even overlooked by existing methods: taking into account the distance between cluster representatives when assessing the similarity of clusters results and constructing a unified framework for defining a distance based on either hard or soft clustering.

EVCLUS: evidential clustering of proximity data

  • T. DenoeuxM. Masson
  • Computer Science
    IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
  • 2004
A notion of credal partition is introduced, which subsumes those of hard, fuzzy, and possibilistic partitions, allowing to gain deeper insight into the structure of the data.

Comparing Fuzzy, Probabilistic, and Possibilistic Partitions

This paper generalizes many of the classical indices that have been used with outputs of crisp clustering algorithms so that they are applicable for candidate partitions of any type (i.e., crisp or soft, with soft comprising the fuzzy, probabilistic, and possibilistic cases).

A possibilistic approach to clustering

An appropriate objective function whose minimum will characterize a good possibilistic partition of the data is constructed, and the membership and prototype update equations are derived from necessary conditions for minimization of the criterion function.

ECM: An evidential version of the fuzzy c

Orthopartitions and soft clustering: Soft mutual information measures for clustering validation

External Indices for Rough Clustering

Clustering external indices are used to compare the clustering result with a given gold standard, represented (in the classical case) by a partition of the dataset. Rough clustering on the other hand

Comparing Fuzzy Partitions: A Generalization of the Rand Index and Related Measures

This paper introduces a fuzzy extension of a class of measures to compare clustering structures, namely, measures that are based on the number of concordant and theNumber of discordant pairs of data points, and exhibits desirable metrical properties.