Transforming a dataset into one with fewer columns is more complicated than it might seem, explains Dr. James McCaffrey of Microsoft Research in this full-code, step-by-step machine learning tutorial.

A very important technique in unsupervised machine learning as well as dimensionality reduction is Principal Component Analysis (PCA). But PCA is difficult to understand without the fundamental ...

Sankhyā: The Indian Journal of Statistics, Series A (1961-2002), Vol. 26, No. 4 (Dec., 1964), pp. 329-358 (30 pages) The paper provides various interpretations of principal components in the analysis ...

We examined the ability of eigenvalue tests to distinguish field-collected from random, assemblage structure data sets. Eight published time series of species abundances were used in the analysis, ...