Dimensionality Reduction for Data Visualization [Applications Corner]

Samuel Kaski, Jaakko Peltonen

Abstract: Dimensionality reduction is one of the basic operations in the toolbox of data analysts and designers of machine learning and pattern recognition systems. Given a large set of measured variables but few observations, an obvious idea is to reduce the degrees of freedom in the measurements by rep resenting them with a smaller set of more "condensed" variables. Another reason for reducing the dimensionality is to reduce computational load in further processing. A third reason is visualization.