A General Approach to Clustering in Large Databases with Noise

Abstract

Several clustering algorithms can be applied to clustering in large multimedia databases. The effectiveness and efficiency of the existing algorithms, however, are somewhat limited, since clustering in multimedia databases requires clustering of high-dimensional feature vectors and because multimedia databases often contain large amounts of noise. In this paper, we therefore introduce a new Kernel Density Estimation-based algorithm for clustering in large multimedia databases called DENCLUE (DENsity-based CLUstEring). Kernel Density Estimation (KDE) models the overall point density analytically as the sum of kernel (or influence) functions of the data points. Clusters can then be identified by determining density attractors and clusters of arbitrary shape can be easily described by a simple equation of the overall density function. The advantages of our KDE-based DENCLUE approach are: (1) it has a firm mathematical basis; (2) it has good clustering properties in data sets with large amounts of noise; (3) it allows a compact mathematical description of arbitrarily shaped clusters in high-dimensional data sets; and (4) it is significantly faster than existing algorithms. To demonstrate the effectiveness and efficiency of DENCLUE, we perform a series of experiments on a number of different data sets from CAD and molecular biology. A comparison with k-Means, DBSCAN, and BIRCH shows the superiority of our new algorithm.