Fuzzy clustering algorithms with distance metric learning and entropy regularization

Clustering has been used in various fields, such as image processing, data mining, pattern recognition, and statistical analysis. Generally, clustering algorithms consider all variables equally relevant or not correlated. Nevertheless, the of samples multidimensional space can be geometrically complicated, e.g., clusters may exist different subsets features. In this regard, new soft subspace have proposed, which correlation relevance are considered to improve their performance. Since regularization-based methods robust for initializations, approaches proposed introduce an entropy regularization term controlling membership degree objects. Such regularizations popular due high performance large-scale low computational complexity. These three-step iterative provide a fuzzy partition, representative each cluster, weight by minimizing suitable objective function. Several experiments on synthetic real datasets, including application segmentation noisy textures, demonstrate usefulness methods.