Standardizing the comparison of partitions

DOI10.1007/S001800300136MaRDI QIDQ1424639zbMATH OpenOpenAlexFDO

Authors Ursula Garczarek, Claus Weihs

Publication date 16 March 2004

Published in Computational Statistics (Search for Journal in Brave)

Full work available at URL http://hdl.handle.net/2003/5274

zbMATH Keywords

accuracyclassifiernon-resemblancevisualization

Mathematics Subject Classification ID

Graphical methods in statistics (62A09) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Review text

A class of \(\arg\max\) classification rules is considered in which a membership function \(m(x,c)\) is attached to each class \(c=1,\dots,G\). An object with predictor value \(x\) is classified to the class \(\hbox{cl}(x)=\arg\max_{c=1,\dots,G} m(x,c)\). (In Bayes classification \(m(x,c)\) are the posterior probabilities to belong to the class \(c\)). The authors propose a standardizing technique for \(m(x,c)\) which allows to compare different \(\arg\max\) classifiers. This technique captures such features of the classifier as non-resemblance (the quality of membership values separation in different classes). Vizualization of the obtained partitions for three (or four) classes is discussed. Results are presented for classification of simulated chi-square distributed r.v.s and for Fisher's Iris data.

Recommendations

• Standardized partition spaces
• scientific article; zbMATH DE number 90337
• Criteria for comparing clusterings
• scientific article; zbMATH DE number 3942813
• Comparing partitions of two sets of units based on the same variables

Cited in

(2)

• Standardized partition spaces
• Partitioning datasets based on equalities among parameters

Describes a project that uses

Uses Software

• BayesDA
• R
• UCI
• Neural Network Toolbox

This page was built for publication: Standardizing the comparison of partitions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1424639)