fuzzy partitions). As our ultimate goal is to compare the results of standard fuzzy clustering algorithms (e.g. fuzzy c-means), we define a method to construct them from a set of fuzzy clusters obtained from several executions of fuzzy c-means. From a practical point of view, the approach presented here tries to solve the difficulty of comparing the results
of fuzzy clustering methods and, in particular, the difficulty of finding the global optimal.
- Content Type Journal Article
- DOI 10.1007/s00500-010-0605-z
- Authors
- Vicenç Torra, CSIC, Spanish Council for Scientific Research IIIA, Institut d’Investigació en Intel-ligència Artificial Campus de Bellaterra 08193 Bellaterra Catalonia Spain
- Sadaaki Miyamoto, University of Tsukuba Department of Risk Engineering, School of Systems and Information Engineering Tsukuba Ibaraki 305-8573 Japan
- Journal Soft Computing – A Fusion of Foundations, Methodologies and Applications
- Online ISSN 1433-7479
- Print ISSN 1432-7643