Abstract
Characterization of fuzzy clustering algorithms is dealt with using the concept of probabilistic sets and their subjective entropy. Clustering techniques with objective performance-index criterion-functions are discussed. Several indices characterized by subjective entropy, including indices expressing the degree of interaction between clusters, are proposed. The utility of these indices is illustrated by a simulation method using Iris data and also handwritten-character data.
| Original language | English |
|---|---|
| Title of host publication | Anal of Fuzzy Inf |
| Publisher | CRC Press Inc |
| Pages | 169-181 |
| Number of pages | 13 |
| ISBN (Print) | 0849362989 |
| Publication status | Published - 1987 |