Advertisement

Hybrid k-Means: Combining Regression-Wise and Centroid-Based Criteria for QSAR

  • Robert Stanforth
  • Evgueni Kolossov
  • Boris Mirkin
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)

Abstract

This paper further extends the ‘kernel’-based approach to clustering proposed by E. Diday in early 70s. According to this approach, a cluster’s centroid can be represented by parameters of any analytical model, such as linear regression equation, built over the cluster. We address the problem of producing regression-wise clusters to be separated in the input variable space by building a hybrid clustering criterion that combines the regression-wise clustering criterion with the conventional centroid-based one.

Keywords

Feature Space Loss Function Hybrid Model Cluster Criterion Relative Prediction Error 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. DIDAY, E. (1974): Optimization in non-hierarchical clustering. Pattern Recognition 6(1), 17–33.CrossRefGoogle Scholar
  2. DIDAY, E., CELEUX, G., GOVAERT, G., LECHEVALLIER, Y., and RALAMBONDRAINY, H. (1989): Classification Automatique des Données. Dunod, Paris.Google Scholar
  3. MIRKIN, B. (2005): Clustering for Data Mining: A Data Recovery Approach. Chapman & Hall/CRC, Boca Raton, Fl.zbMATHGoogle Scholar
  4. TABACHNICK, B.G. and FIDELL, L.S. (2006): Using Multivariate Statistics (5th Edition). Allyn & Bacon, Boston.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Robert Stanforth
    • 1
    • 2
  • Evgueni Kolossov
    • 1
  • Boris Mirkin
    • 2
  1. 1.ID Business SolutionsGuildfordUK
  2. 2.School of Computer Science, BirkbeckUniversity of LondonLondonUK

Personalised recommendations