CONVENTIONAL AND BAYESIAN VALIDATION FOR FUZZY CLUSTERING ANALYSIS

Olfa Limam, Fouad Ben Abdelaziz

2010

Abstract

Clustering analysis has been used for identifying similar objects and discovering distribution of patterns in large data sets. While hard clustering assigns an object to only one cluster, fuzzy clustering assigns one object to multiple clusters at the same time based on their degrees of membership. An important issue in clustering analysis is the validation of fuzzy partitions. In this paper, we consider the Bayesian like validation along with four conventional validity measures for two clustering algorithms namely, fuzzy c-means and fuzzy c-shell based. An empirical study is conducted on five data sets to compare their performances. Results show that the Bayesian validation score outperforms the conventional ones. However, a multiple objective approach is needed.

References

  1. Bezdek, J. (1974). Cluster validity with fuzzy sets. Journal of Cybernetics and Systems, 3(3):58-72.
  2. Carvalho, F. (2006). A fuzzy clustering algorithm for symbolic interval data based on a single adaptive euclidean distance. In ICONIP (3), pages 1012-1021.
  3. Cho, S. and Yoo, S. (2006). Fuzzy bayesian validation for cluster analysis of yeast cell-cycle data. Pattern Recognition, 39(12):2405-2414.
  4. Dave, R. (1990). Fuzzy shell-clustering and applications to circle detection in digital image. International Journal of General Systems, 16(12):343-355.
  5. Fukuyama, Y. and Suengo, M. (1989). A new method of choosing the number of clusters for the fuzzy c-means. Proceedings of Fifth Fuzzy Systems Symposium, pages 247-250.
  6. Graves, D. and Pedrycz, W. (2010). Kernel-based fuzzy clustering and fuzzy clustering: A comparative experimental study. Fuzzy Sets and Systems, 161(4):522- 543.
  7. Halkidi, M., Batistakis, Y., and M.Vazirgiannis (2001). On clustering validation techniques. J. Intell. Inf. Syst., 17(2-3):107-145.
  8. Klawonn, F. and Hoppne, F. (2009). Fuzzy cluster analysis from the viewpoint of robust statistics. Studies in Fuzziness and Soft Computing, 243(19):439-455.
  9. Wang, W. and Zhang, Y. (2007). On fuzzy cluster validity indices. Fuzzy Sets and Systems, 158(19):2095-2117.
  10. Xie, X. and Beni, G. (1991). A validity measure for fuzzy clustering. IEEE Trans. Pattern Anal. Mach. Intell., 13(8):841-847.
  11. Yang, X., Cao, A., and Song, Q. (2006). A new cluster validity for data clustering. Neural Processing Letters, 23(3):325-344.
Download


Paper Citation


in Harvard Style

Limam O. and Ben Abdelaziz F. (2010). CONVENTIONAL AND BAYESIAN VALIDATION FOR FUZZY CLUSTERING ANALYSIS . In Proceedings of the International Conference on Fuzzy Computation and 2nd International Conference on Neural Computation - Volume 1: ICFC, (IJCCI 2010) ISBN 978-989-8425-32-4, pages 135-140. DOI: 10.5220/0003110201350140


in Bibtex Style

@conference{icfc10,
author={Olfa Limam and Fouad Ben Abdelaziz},
title={CONVENTIONAL AND BAYESIAN VALIDATION FOR FUZZY CLUSTERING ANALYSIS},
booktitle={Proceedings of the International Conference on Fuzzy Computation and 2nd International Conference on Neural Computation - Volume 1: ICFC, (IJCCI 2010)},
year={2010},
pages={135-140},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003110201350140},
isbn={978-989-8425-32-4},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Fuzzy Computation and 2nd International Conference on Neural Computation - Volume 1: ICFC, (IJCCI 2010)
TI - CONVENTIONAL AND BAYESIAN VALIDATION FOR FUZZY CLUSTERING ANALYSIS
SN - 978-989-8425-32-4
AU - Limam O.
AU - Ben Abdelaziz F.
PY - 2010
SP - 135
EP - 140
DO - 10.5220/0003110201350140