Cross-entropy clustering

DOI10.1016/J.PATCOG.2014.03.006zbMATH Open1342.68279arXiv1210.5594OpenAlexW2063598178WikidataQ112225956 ScholiaQ112225956MaRDI QIDQ85407FDOQ85407

Jacek Tabor, Przemysław Spurek, Jacek Tabor, Przemysław Spurek

Publication date: September 2014

Published in: Pattern Recognition (Search for Journal in Brave)

Abstract: We construct a cross-entropy clustering (CEC) theory which finds the optimal number of clusters by automatically removing groups which carry no information. Moreover, our theory gives simple and efficient criterion to verify cluster validity. Although CEC can be build on an arbitrary family of densities, in the most important case of Gaussian CEC: {em -- the division into clusters is affine invariant; -- the clustering will have the tendency to divide the data into ellipsoid-type shapes; -- the approach is computationally efficient as we can apply Hartigan approach.} We study also with particular attention clustering based on the Spherical Gaussian densities and that of Gaussian densities with covariance $sI$. In the letter case we show that with $s$ converging to zero we obtain the classical k-means clustering.

Full work available at URL: https://arxiv.org/abs/1210.5594

Recommendations

• Application of the cross-entropy method to clustering and vector quantization
• scientific article; zbMATH DE number 1098881
• Entropic approach to multiscale clustering analysis
• A clustering model with Rényi entropy regularization
• scientific article; zbMATH DE number 1975248
• K-MEANS CLUSTERING USING ENTROPY MINIMIZATION

zbMATH Keywords

clusteringcross-entropymemory compression

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Pattern recognition, speech recognition (68T10)

Cites Work

• Estimating the number of clusters in a data set via the gap statistic
• Title not available (Why is that?)
• Title not available (Why is that?)
• Model-Based Gaussian and Non-Gaussian Clustering
• Least squares quantization in PCM
• NP-hardness of Euclidean sum-of-squares clustering
• Competitive EM algorithm for finite mixture models
• Applied Multivariate Statistical Analysis
• Title not available (Why is that?)
• Title not available (Why is that?)
• Bayesian k-Means as a “Maximization-Expectation” Algorithm
• Title not available (Why is that?)
• Printer graphics for clustering
• A deterministic annealing approach to clustering
• Title not available (Why is that?)
• Title not available (Why is that?)
• Clustering Methods: A History of k-Means Algorithms
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• On clustering validation techniques

Cited In (13)

• Clustering of seasonal events: A simulation study using circular methods
• R2DS: a novel hierarchical framework for driver fatigue detection in mountain freeway
• Ellipticity and circularity measuring via Kullback-Leibler divergence
• Extreme entropy machines: robust information theoretic classification
• Constrained clustering with a complex cluster structure
• Lossy compression approach to subspace clustering
• K-MEANS CLUSTERING USING ENTROPY MINIMIZATION
• Neighborhood density information in clustering
• CEC
• Logistic regression with weight grouping priors
• Semi-supervised cross-entropy clustering with information bottleneck constraint
• Title not available (Why is that?)
• Efficient mixture model for clustering of sparse high dimensional binary data

Uses Software

• EDA
• k-means++

This page was built for publication: Cross-entropy clustering

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q85407)