Robust Extension of FCM Algorithm

Autor: Cheng-Jia Li
Rok vydání: 2006
Předmět:
Zdroj: 2006 International Conference on Machine Learning and Cybernetics.
DOI: 10.1109/icmlc.2006.258710
Popis: Clustering is a procedure through which objects are distinguished or classified in accordance with their similarity. The fuzzy c-means method (FCM) is one of the most popular clustering methods based on minimization of a criterion function. However, the FCM method is sensitive to the presence of noise and outliers in data. A new clustering algorithm is proposed by extending the criterion function, which includes the well-known fuzzy c-means method as its special case. Numerical experiments show that the new clustering algorithm is less sensitive than the traditional FCM method and robust to outliers.
Databáze: OpenAIRE