Criterion-based divisive clustering for symbolic data

MaRDI QIDQ2756650zbMATH OpenFDO

Authors Marie Chavent

Publication date 18 November 2001

zbMATH Keywords

clusteringdata arraysdendrogramshierarchiessymbolic data

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) General topics in the theory of data (68P01)

Review text

A divisive clustering method for a symbolic \(n\times p\) data array \(\underline{X}=\{Y_j(k)\}\) is proposed. In the proposed clustering method a division is defined by optimizing a suitable generalization of the classical within-clusters variance criterion to the case of symbolic data. This criterion is then minimized with respect to all the bipartitions induced by a set of binary questions relating to any of the symbolic variables \(Y_j\) with an ordered domain \({\mathcal Y}_j\) (observation space) such that the data array may contain classical, multivalent, interval or modal variables. The clusters are not systematically split, but one of the clusters is chosen according to a specific criterion. The splitting process is stopped after a number of iterations which may be specified by the user who is typically interested in a classification with a small number of clusters. The output of this divisive clustering method is an indexed hierarchy (dendrogram) which is also a decision tree. For the entire collection see [Zbl 1039.62501].

Recommendations

• scientific article; zbMATH DE number 6902608
• A monothetic clustering method
• Hierarchical and pyramidal clustering for symbolic data
• Dissimilarity measures and divisive clustering for symbolic multimodal-valued data
• DIVCLUS-T: a monothetic divisive hierarchical clustering method

Cited in

(12)

• Partitional clustering algorithms for symbolic interval data based on single adaptive distances
• Dissimilarity measures and divisive clustering for symbolic multimodal-valued data
• Monothetic divisive clustering with geographical constraints
• Clustering and its validation in a symbolic framework
• scientific article; zbMATH DE number 6902608 (Why is no real title available?)
• scientific article; zbMATH DE number 2101171 (Why is no real title available?)
• An Agglomerative Hierarchical Clustering Algorithm for Improving Symbolic Object Retrieval
• Hierarchical and pyramidal clustering for symbolic data
• Fuzzy K-means clustering algorithms for interval-valued data based on adaptive quadratic distances
• A robust fuzzy k-means clustering model for interval valued data
• Clustering reduced interval data using Hausdorff distance
• A polythetic clustering process and cluster validity indexes for histogram-valued objects

This page was built for publication: Criterion-based divisive clustering for symbolic data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2756650)