STEPWISE ALGORITHM FOR FINDING UNKNOWN NUMBER OF CLUSTERS

@inproceedings{Krkkinen2002STEPWISEAF,
  title={STEPWISE ALGORITHM FOR FINDING UNKNOWN NUMBER OF CLUSTERS},
  author={Ismo K{\"a}rkk{\"a}inen and Pasi Fr{\"a}nti},
  year={2002}
}
We consider a criterion-based approach to solve dynamic clustering problems. We propose two novel improvements for reducing the work load to find the correct clustering. The first idea is to use stepwise clustering algorithm that utilizes the previous solution when solving the next clustering with different (one more or one less) number of clusters. The second idea is to use a heuristic stopping criterion in the algorithm that solves a single clustering. The stopping criterion estimates… CONTINUE READING

Citations

Publications citing this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 25 REFERENCES

A Cluster Separation Measure

  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 1979
VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

Minimization of the value of Davies- Bouldin index

I Kärkkäinen, P Fränti
  • In Proceedings of the IASTED International Conference on Signal Processing and Communications (SPC’2000). IASTED/ACTA Press,
  • 2000
VIEW 2 EXCERPTS

Some new indexes of cluster validity

  • IEEE Trans. Systems, Man, and Cybernetics, Part B
  • 1998
VIEW 2 EXCERPTS