Speed Session: Clustering Singular and Non-Singular Covariance Matrices for Classification
Andrew Simpson is a Ph.D. Student in the Computational Science and Statistics program at South Dakota State University. Research focuses on novel methods for modeling data generated from a hierarchical sampling process where subpopulation structures exist. The main application of this research is to forensic statistics and source identification.
In classification problems when working in high dimensions with a large number of classes and few observations per class, linear discriminant analysis (LDA) requires the strong assumptions of a shared covariance matrix between all classes and quadratic discriminant analysis leads to singular or unstable covariance matrix estimates. Both of these can lead to lower than desired classification performance. We introduce a novel, model-based clustering method which can relax the shared covariance assumptions of LDA by clustering sample covariance matrices, either singular or non-singular. This will lead to covariance matrix estimates which are pooled within each cluster. We show using simulated and real data that our method for classification tends to yield better discrimination compared to other methods.
Session Materials: https://dataworks.testscience.org/wp-...