A Robust Competitive Clustering Algorithm With Applications in Computer Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
Towards a robust fuzzy clustering
Fuzzy Sets and Systems - Data analysis
A Similarity-Based Robust Clustering Method
IEEE Transactions on Pattern Analysis and Machine Intelligence
Unsupervised possibilistic clustering
Pattern Recognition
A novel fuzzy clustering algorithm based on a fuzzy scatter matrix with optimality tests
Pattern Recognition Letters
Analysis of the weighting exponent in the FCM
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Alpha-Cut Implemented Fuzzy Clustering Algorithms and Switching Regressions
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Robust clustering methods: a unified view
IEEE Transactions on Fuzzy Systems
Will the real iris data please stand up?
IEEE Transactions on Fuzzy Systems
Optimality test for generalized FCM and its application to parameter selection
IEEE Transactions on Fuzzy Systems
A fuzzy-logic-based approach to qualitative modeling
IEEE Transactions on Fuzzy Systems
On cluster validity for the fuzzy c-means model
IEEE Transactions on Fuzzy Systems
Mathematical and Computer Modelling: An International Journal
Fuzzy algorithms for combined quantization and dithering
IEEE Transactions on Image Processing
Objective function-based clustering
Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery
Visual data mining for identification of patterns and outliers in weather stations' data
IDEAL'12 Proceedings of the 13th international conference on Intelligent Data Engineering and Automated Learning
Generalized agglomerative fuzzy clustering
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part III
Soft clustering -- Fuzzy and rough approaches and their extensions and derivatives
International Journal of Approximate Reasoning
A multivariate fuzzy c-means method
Applied Soft Computing
Hi-index | 0.01 |
The weighting exponent m is called the fuzzifier that can influence the performance of fuzzy c-means (FCM). It is generally suggested that m@?[1.5,2.5]. On the basis of a robust analysis of FCM, a new guideline for selecting the parameter m is proposed. We will show that a large m value will make FCM more robust to noise and outliers. However, considerably large m values that are greater than the theoretical upper bound will make the sample mean a unique optimizer. A simple and efficient method to avoid this unexpected case in fuzzy clustering is to assign a cluster core to each cluster. We will also discuss some clustering algorithms that extend FCM to contain the cluster cores in fuzzy clusters. For a large theoretical upper bound case, we suggest the implementation of the FCM with a suitable large m value. Otherwise, we suggest implementing the clustering methods with cluster cores. When the data set contains noise and outliers, the fuzzifier m=4 is recommended for both FCM and cluster-core-based methods in a large theoretical upper bound case.