Machine Learning with R Cookbook - Second Edition

Performing nonlinear dimension reduction with Local Linear Embedding

Local Linear Embedding (LLE) is an extension of PCA, which reduces data that lies on a manifold embedded in a high dimensional space into a low dimensional space. In contrast to ISOMAP, which is a global approach for nonlinear dimension reduction, LLE is a local approach that employs a linear combination of the k-nearest neighbor to preserve local properties of data. In this recipe, we will give a short introduction of how to use LLE for s-curve data.