Locally linear embedding for nonlinear dimension reduction in classification problems: an application to gene expression data

Some real problems, such as image recognition or the analysis of gene expression data, involve the observation of a very large number of variables on a few units. In such a context conventional classification methods are difficult to employ both from analytical and interpretative points of view. In this paper, a solution based on locally linear embedding (LLE) for supervised classification is proposed and applied to the analysis of some gene expression datasets.