|
Abstract:
|
Cluster analysis is a common technique for the identification of groups of observations that are cohesive and separated from other groups, which is being applied to a variety of areas, such as biology, market research, image analysis, social network analysis, and data mining. Model-based clustering arose from the multivariate normal mixture models with covariances parameterized by eigenvalue decomposition, which is hierarchical agglomeration based on the classification likelihood and the expectation-maximization algorithm for maximum likelihood estimation of multivariate mixture models. The purpose of this study is to review the key concepts on clustering methods and illustrate how to apply various clustering methods in data exploration with an emphasis on model-based clustering. Mike's food and nutrition data was used to demonstrate the application of model-based clustering method.
|
