Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

A Fast K-means and Fuzzy-c-means Algorithms using Adaptively Initialization

적응적인 초기치 설정을 이용한 Fast K-means 및 Frizzy-c-means 알고리즘

  • Published : 2004.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, the initial value problem in clustering using K-means or Fuzzy-c-means is considered to reduce the number of iterations. Conventionally the initial values in clustering using K-means or Fuzzy-c-means are chosen randomly, which sometimes brings the results that the process of clustering converges to undesired center points. The choice of intial value has been one of the well-known subjects to be solved. The system of clustering using K-means or Fuzzy-c-means is sensitive to the choice of intial values. As an approach to the problem, the uniform partitioning method is employed to extract the optimal initial point for each clustering of data. Experimental results are presented to demonstrate the superiority of the proposed method, which reduces the number of iterations for the central points of clustering groups.

본 논문에서는 K-means 또는 Fuzzy-c-means 알고리즘에서 클러스터의 중심점을 찾는 과정 중 임의로 선택되는 초기값 선정의 문제를 해결하고, 기존의 단점을 보완하는 새로운 방안으로서 데이터의 분포의 통계적 특성에 따른 초기값 선정 방법을 제안하였다. 기존의 초기값 선정 방법은 초기값에 따라 클러스터링이 매우 민감한 변화를 가져와, 최종적으로 종종 원치 않는 방향으로 가는 문제점을 갖고 있다. 이러한 초기값 선정의 문제가 인지되어 왔지만, 그 문제의 해결방안이 실제적으로 모색된 경우는 없었다. 본 논문에서는 데이타의 통계적 특성을 이용한 초기값 선정 방법을 적용하여, 클러스터링이 형성되는 시간의 단축 및 원치 않는 결과가 생성되는 경우를 약화시켜 시스템의 향상을 가져왔고, 이러한 제안된 알고리즘의 우수성을 기존의 알고리즘과 비교를 통하여 나타내었다.

Keywords

References

  1. C. N. Schizas and C. S. Pattichis, 'Neural networks, genetic algorithms and the K-means algorithm: in search of data classification,' COGANN-92. International Workshop, pp. 201-222, Jun 1992
  2. R. N. Dave, 'Robust fuzzy clustering algorithms,' IEEE Fuzzy Systems International Conference, vol. 2, pp. 1281-1286, 1993 https://doi.org/10.1109/FUZZY.1993.327577
  3. G. Beni and Liu Xiamoin, 'A least biased fuzzy clustering method,' IEEE Pattern Analysis and Machine Intelligence Trans, vol. 16, pp, 954-960, Sep 1994 https://doi.org/10.1109/34.310694
  4. C.M. Bishop, Neural Networks for Pattern Recognition, Oxford University Press, 1995
  5. M. Singh and P. Patel, and D. Khosla, 'Segmentation of functional MRI by K -means clustering,' IEEE Nuclear Science Symposium and Medical Imaging Conference, vol. 3, pp. 1732-1736, Oct 1995 https://doi.org/10.1109/NSSMIC.1995.501920
  6. Timothy J. Ross, Fuzzy Logic with Engineering Application, McGraw-HilI, Inc., 1995
  7. R. Krishnapuram and J. Kim 'Clustering algorithms based on volume criteria,' IEEE Fuzzy Systems Trans, vol. 8, pp. 228-236, Apr 2000 https://doi.org/10.1109/91.842156
  8. Wong Ching-Chang and Chen Chia-Chong, 'K-means-based fuzzy classifier design,' IEEE Fuzzy Systems International Conference, vol. 1, pp, 48-52, May 2000 https://doi.org/10.1109/FUZZY.2000.838632
  9. K. K. Paliwal and V. Ramasubramainan, 'Modified K-means algorithm for vector quantizer design,' IEEE Image Processing Trans, vol. 9 pp. 1964-1967, Nov 2000 https://doi.org/10.1109/97.551685
  10. S. S. R. Abicli and J. Ong, 'A data mining strategy for inductive data clustering: a synergy between self-organising neural networks and K-means clustering techniques,' TENCON 2000. Proceedings , vol. 2, pp. 568-573, 2000 https://doi.org/10.1109/TENCON.2000.888802
  11. Ian T. Nabney, NETLAB Algorithms for Pattern Recognition, Springer, 2001
  12. R.O. Duda and P.E. Hart, Pattern Classification, Willey, 2001
  13. Su Mu-Chun and Chou Chien-Hsing, 'A modified version of the K-means algorithm with a distance based on cluster symmetry,' IEEE Pattern Analysis and Machine Intelligence Trans, vol. 23, pp. 674-680, Jun 2001 https://doi.org/10.1109/34.927466