ABSTRACT
This chapter explains in what ways unsupervised learning can be used, especially in the data pre-processing phase. It presents two approaches that help reduce the number of predictors. The first one aims at creating new variables that are uncorrelated with each other. Low correlation is favorable from an algorithmic point of view, but the new variables lack interpretability. The second one gathers predictors into homogeneous clusters and only one feature should be chosen out of this cluster. The rationale is reversed: interpretability is favored over statistical properties because the resulting set of features may still include high correlations, albeit to a- lesser point compared to the original one. An autoencodeur generalizes this concept to nonlinear coding functions. Simple linear autoencoders are linked to latent factor models for the case of single layer autoencoders.
