Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Special feature: dimension reduction and cluster analysis

Special feature: dimension reduction and cluster analysis Behaviormetrika https://doi.org/10.1007/s41237-019-00092-6 PREFACE 1 2 3 Michel van de Velden  · Alfonso Iodice D’Enza  · Michio Yamamoto © The Behaviormetric Society 2019 Dimension reduction and cluster analysis have a long history in multivariate data analysis. Dimension reduction methods typically concern themselves with a reduc- tion in the variable space through either selection of variables or the construction of new variables as combinations of the original ones. Cluster analysis aims to detect groups of similar observations thus reducing the row space. Although these meth- ods typically consider different objective functions, the ultimate goals, e.g., detect- ing and summarizing relevant properties and relationships in the data, are typically similar if not identical. Consequently, dimension reduction and cluster analysis are frequently combined. One way to combine dimension reduction and cluster analysis is to perform the analyses sequentially. In particular, a common procedure is to first perform dimen- sion reduction and then apply cluster analysis to the reduced data. However, Bock (1987), Van  Buuren and Heiser (1989) and De  Soete and Carroll (1994), already observed that such sequential analyses may not be optimal due to the differences between the objective functions corresponding to the dimension reduction and cluster analysis parts. An alternative to such sequential methods, is to formulate http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behaviormetrika Springer Journals

Special feature: dimension reduction and cluster analysis

Velden, Michel; D’Enza, Alfonso; Yamamoto, Michio
Behaviormetrika , Volume OnlineFirst – Sep 13, 2019
Download PDF
3 pages

Loading next page...
 
/lp/springer-journals/special-feature-dimension-reduction-and-cluster-analysis-ZNUHiTix6J
Publisher
Springer Journals
Copyright
Copyright © 2019 by The Behaviormetric Society
Subject
Statistics; Statistical Theory and Methods; Statistics for Business, Management, Economics, Finance, Insurance; Statistics for Engineering, Physics, Computer Science, Chemistry and Earth Sciences
ISSN
0385-7417
eISSN
1349-6964
DOI
10.1007/s41237-019-00092-6
Publisher site
See Article on Publisher Site

Abstract

Behaviormetrika https://doi.org/10.1007/s41237-019-00092-6 PREFACE 1 2 3 Michel van de Velden  · Alfonso Iodice D’Enza  · Michio Yamamoto © The Behaviormetric Society 2019 Dimension reduction and cluster analysis have a long history in multivariate data analysis. Dimension reduction methods typically concern themselves with a reduc- tion in the variable space through either selection of variables or the construction of new variables as combinations of the original ones. Cluster analysis aims to detect groups of similar observations thus reducing the row space. Although these meth- ods typically consider different objective functions, the ultimate goals, e.g., detect- ing and summarizing relevant properties and relationships in the data, are typically similar if not identical. Consequently, dimension reduction and cluster analysis are frequently combined. One way to combine dimension reduction and cluster analysis is to perform the analyses sequentially. In particular, a common procedure is to first perform dimen- sion reduction and then apply cluster analysis to the reduced data. However, Bock (1987), Van  Buuren and Heiser (1989) and De  Soete and Carroll (1994), already observed that such sequential analyses may not be optimal due to the differences between the objective functions corresponding to the dimension reduction and cluster analysis parts. An alternative to such sequential methods, is to formulate

Journal

Behaviormetrika – Springer Journals

Published: Sep 13, 2019

References

  • Probabilistic d-clustering
    Ben-Israel, A; Iyigun, C
  • Multivariate statistical modeling and data analysis
    Bock, H
  • New approaches in classification and data analysis
    Soete, G; Carroll, JD
  • An extension of multiple correspondence analysis for identifying heterogeneous subgroups of respondents
    Hwang, H; Dillon, WR; Takane, Y
  • Analyzing and modeling rank data
    Marden, JI
  • Clustering n objects into k groups under optimal scaling of variables
    Buuren, S; Heiser, WJ
  • Cluster correspondence analysis
    Velden, M; Iodice D’Enza, A; Palumbo, F
  • Factorial $$k$$ k -means analysis for two-way data
    Vichi, M; Kiers, H
  • A general formulation of cluster analysis with dimension reduction and subspace separation
    Yamamoto, M; Hwang, H
  • Dimension-reduced clustering of functional data via subspace separation
    Yamamoto, M; Hwang, H