Abstract
The 'fuzzy clustering' problem is investigated. Interesting properties of the points generated in the course of applying the fuzzy c-means algorithm are revealed using the concept of reduced objective function. We investigate seven quantities that could be used for stopping the algorithm and prove relationships among them. Finally, we empirically show that these quantities converge linearly.
| Original language | English |
|---|---|
| Pages (from-to) | 181-191 |
| Number of pages | 11 |
| Journal | Fuzzy Sets and Systems |
| Volume | 49 |
| Issue number | 2 |
| DOIs | |
| State | Published - 27 Jul 1992 |
| Externally published | Yes |
Bibliographical note
Funding Information:This work was partially supported by the Natural Science and Engineering Research Council of Canada through a research grant to the second author.
Keywords
- Fuzzy c-means algorithm
- convergence of fuzzy c-means algorithm
- fuzzy clustering
- stopping criteria for fuzzy c-means
ASJC Scopus subject areas
- Logic
- Artificial Intelligence