Model-Based Clustering

In this chapter, we present a method for clustering that is based on finite mixture probabil i ty models . We f irs t provide an overview of a comprehensive procedure for model-based clustering, so the reader has a framework for the methodology. We then describe the constituent techniques used in model-based clustering, such as finite mixtures, the EM algorithm, and model-based agglomerative clustering. We then put it all together again and include more discussion on its use in EDA and clustering. Finally, we show how to use a GUI tool that generates random samples based on the finite mixture models presented in the chapter.