An Overview of Unsupervised and Semi-Supervised Fuzzy Kernel Clustering

摘要: For real-world clustering tasks, the input data is typically not easily separable due to highly complex structure or when clusters vary in size, density and shape. Kernel-based has proven be an effective approach partition such data. In this paper, we provide overview of several fuzzy kernel algorithms. We focus on methods that optimize C-mean-type objective function. highlight advantages disadvantages each method. addition completely unsupervised algorithms, also some semi-supervised These algorithms use partial supervision information guide optimization process avoid local minima. different approaches have been used extend handle very large sets.