Self-updating Clustering Algorithm for Probability Density Functions

Abstract

In this talk, we propose an automatic clustering algorithm, called self-updating clustering algorithm, for grouping the probability density functions. To determine the suitable widths of the clusters, a data-driven learning mechanism is incorporated in the algorithm based on the learning of Self-Organizing Map. Experimental results confirm the effectiveness of the proposed algorithm when applied to probability density functions. In addition, the clustering can serve as the intermediate compression tool in content-based multimedia retrieval that we apply the proposed algorithm to categorize a subset of COREL image database. And the clustering results indicate that the proposed algorithm performs well in color image categorization.