A Two-Stage Method for Near-Optimal Clustering

최적에 가까운 군집화를 위한 이단계 방법

  • 윤복식 (홍익대학교 기초과학과 응용수학전공)
  • Published : 2004.03.01

Abstract

The purpose of clustering is to partition a set of objects into several clusters based on some appropriate similarity measure. In most cases, clustering is considered without any prior information on the number of clusters or the structure of the given data, which makes clustering is one example of very complicated combinatorial optimization problems. In this paper we propose a general-purpose clustering method that can determine the proper number of clusters as well as efficiently carry out clustering analysis for various types of data. The method is composed of two stages. In the first stage, two different hierarchical clustering methods are used to get a reasonably good clustering result, which is improved In the second stage by ASA(accelerated simulated annealing) algorithm equipped with specially designed perturbation schemes. Extensive experimental results are given to demonstrate the apparent usefulness of our ASA clustering method.

References

  1. 적절한 군집수 결정방법(준비중) 강금석;윤복식
  2. 메타 휴리스틱 김여근;윤복식;이상복
  3. 한국경영과학회지 v.20 no.2 Simulated Annealing의 가속화와 ATM 망에서의 가상경로 설정 윤복식;조계연
  4. Pattern Classification Duda,R.O.;P.E.Hart;D.G.Hart
  5. Cluster analysis Everitt,B.S.
  6. Applied Multivariate Data Analysis Everitt,B.S.
  7. Annals of Eugenics v.2 The use of multiple measurements in taxonomic problems Fisher,R.A.
  8. Discrimination and Classification Hand,D.J.
  9. Science v.220 Optimization by simulated annealing Kirkpatrick,S.;C.D.Gelatt;M.P.Vecchi https://doi.org/10.1126/science.220.4598.671
  10. Pattern Recognition Letters v.22 Stochastic K-means algorithm for vector quantization Kovesi,B.;Boucher,J.M.;Saoudi,S.c https://doi.org/10.1016/S0167-8655(01)00021-6
  11. Multivariate Statistical Methods (Second Ed.) Manly,B.J.
  12. Mathematical Classification and Clustering Mirkin,B.
  13. Journal of Information Sciences v.109 Simulated annealing based pattern classification Sanghamitra,B.;S.K Pal;C.A.Murthy https://doi.org/10.1016/S0020-0255(98)00017-6
  14. International Journal of Pattern Recognition and Artificial Intelligence v.15 no.2 Clustering using simulated annealing with probabilistic redistribution Sanghamitra,B.M.Ujjwal;K.P.Malay https://doi.org/10.1142/S0218001401000927
  15. IEEE Trans. Patt. Anal. Mach. Intell v.6 K-mean type algorithms : a generalized convergence theorem and characterization of local optimality Selim,S.Z.;M.A.Ismail https://doi.org/10.1109/TPAMI.1984.4767478