1. Science
  2. Видання
  3. Системи обробки інформації
  4. 5(86)'2010
  5. Research of dynamic clustering on linearly inseparable plagued experimental data

Research of dynamic clustering on linearly inseparable plagued experimental data

N. Lesna, T. Shatovska, A. Lyakhovets
Системи обробки інформації. — 2010. — № 5(86). – С. 86-91.
UDK 519.6
Article language: english
Annotations languages:


Annotation: The system is representing hierarchical multilevel approach of clustering and explores dynamic modeling in hierarchical clustering. Presented approach operates on a k-nn graph and hyper graph in which nodes represent data items, and weighted edges represent similarities among the data items. Presented hierarchical multilevel algorithm consists of several stages: building the graph, coarsening, partitioning, uncoarsening. Exploring of different combinations of algorithms on different stages for different data sets is the main goal of the work.


Keywords: cluster analysis, hierarchical clustering, coarsening, partitioning, uncoarsening, Chameleon
Information about the authors of publication:
Reference:
Lesna, N.S., Shatovska, T.B. and Lyakhovets, A.V. (2010), Research of dynamic clustering on linearly inseparable plagued experimental data, Information Processing Systems, Vol. 5(86), pp. 86-91.