Abstract
A fuzzy c-means approach to the expectation-maximization (EM) algorithm is proposed in this paper. A family of fuzzy EM algorithms of various degrees of fuzziness is presented, where the EM algorithm is referred to as a fuzzy EM algorithm of degree of fuzziness of one. This fuzzy approach can apply to EM-style algorithms such as the Baum-Welch algorithm for hidden Markov models, the EM algorithm for Gaussian mixture models in speech and speaker recognition. The fuzzy EM algorithm for Gaussian mixture models is considered in detail as a demonstration for applying the fuzzy EM algorithm.
| Original language | English |
|---|---|
| Pages | 421-425 |
| Number of pages | 5 |
| Publication status | Published - 1999 |
| Event | Proceedings of the 1999 18th International Conference of the North American Fuzzy Information Processing Society - NAFIPS'99 - New York, NY, USA Duration: 10 Jun 1999 → 12 Jun 1999 |
Conference
| Conference | Proceedings of the 1999 18th International Conference of the North American Fuzzy Information Processing Society - NAFIPS'99 |
|---|---|
| City | New York, NY, USA |
| Period | 10/06/99 → 12/06/99 |