Towards hierarchical clustering

  • Authors:
  • Mark Sh. Levin

  • Affiliations:
  • Inst. for Information Transmission Problems, Russian Academy of Sciences, Moscow, Russia

  • Venue:
  • CSR'07 Proceedings of the Second international conference on Computer Science: theory and applications
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the paper, new modified agglomerative algorithms for hierarchical clustering are suggested. The clustering process is targeted to generating a cluster hierarchy which can contain the same items in different clusters. The algorithms are based on the following additional operations: (i) building an ordinal item pair proximity ('distance') including the usage of multicriteria approaches; (ii) integration of several item pair at each stage of the algorithms; and (iii) inclusion of the same items into different integrated item pairs/clusters. The suggested modifications above are significant from the viewpoints of practice, e.g., design of systems architecture for engineering and computer systems.