Supervised Clustering: Algorithms and Applications

This work centers on a novel data mining technique we term supervised clustering. Unlike traditional clustering, supervised clustering assumes that the examples are classified and has the goal of identifying class-uniform clusters that have high probability densities. Three representative–based algorithms for supervised clustering are introduced: two greedy algorithms SRIDHCR and SPAM, and an evolutionary computing algorithm named SCEC. The three algorithms were evaluated using a benchmark consisting of UCI machine learning datasets. Experimental results suggest that SCEC outperforms the other two algorithms for almost all data sets in the benchmark. Moreover, a study of the fitness function used by supervised clustering shows that the landscape seems to have a “Canyonland” shape with many hills and plateaus, thereby, increasing the difficulty of the clustering task for the greedy algorithms. Potential applications of supervised clustering are discussed as well. We discuss how supervised clustering can be used for class decomposition and demonstrate with experimental results how it enhances the performance of simple classifiers. We also present a dataset editing technique, we call supervised clustering editing (SCE), which replaces examples of a learned cluster by the cluster representative. Our experimental results demonstrate how dataset editing techniques in general, and SCE technique in particular, enhance the performance of NN classifiers. We also discuss how supervised clustering could be used for regional learning.