Open Conference Systems, 50th Scientific meeting of the Italian Statistical Society

Font Size: 
Some remarks on the DESPOTA algorithm for detecting a partition on a dendrogram
Dario Bruzzese, Domenico Vistocco

Last modified: 2018-06-04

Abstract


The output of hierarchical clustering methods is typically displayed as a dendrogram describing a family of nested partitions. However, the typical approach, horizontally cutting the dendrogram at a given distance level, explores only a restricted subset of the whole set of partitions. DESPOTA (DEndogram Slicing through a PermutatiOn Test Approach) is a novel approach exploiting permutation tests in order to automatically detect a partition among those embedded in a dendrogram. Unlike the traditional approach, DESPOTA includes in the search space also partitions not corresponding to horizontal cuts of the dendrogram.

References


1. Bruzzese, D., Vistocco, D.: DESPOTA: DEndrogram Slicing through a PemutatiOn Test Approach. Journal of Classiï¬cation 32(2), 285–304 (2015)

2. Good, P.I.: Permutations Tests for Testing Hypotheses. Springer-Verlag, New York. (1994)

3. Everitt, B., Landau, M., Leese, M.: Cluster analysis, Arnold, London, 4th edition (2001)

4. ROMANO, J.P., SHAIKH, A.M., and WOLF, M. (2008), “Formalized Data Snooping Based on Generalized Error Ratesâ€, Econometric Theory, 24, 404–447.

5. HOCHBERG, Y. (1988), “A Sharper Bonferroni Procedure for Multiple Tests of Signiï¬canceâ€, Biometrika, 75, 800–802.

6. HOLM, S. (1979), “A Simple Sequentially Rejective Multiple Testing Procedureâ€, Scandinavian Journal of Statistics, 6, 65–70.


Full Text: DOC