Feature Selection
Feature selection approaches try to find a subset of the original variables (also called features or attributes). Two strategies are filter (e.g. information gain) and wrapper (e.g. search guided by the accuracy) approaches. See also combinatorial optimization problems.
In some cases, data analysis such as regression or classification can be done in the reduced space more accurately than in the original space.
Read more about this topic: Dimension Reduction
Famous quotes containing the words feature and/or selection:
“The paid wealth which hundreds in the community acquire in trade, or by the incessant expansions of our population and arts, enchants the eyes of all the rest; the luck of one is the hope of thousands, and the bribe acts like the neighborhood of a gold mine to impoverish the farm, the school, the church, the house, and the very body and feature of man.”
—Ralph Waldo Emerson (1803–1882)
“It is the highest and most legitimate pride of an Englishman to have the letters M.P. written after his name. No selection from the alphabet, no doctorship, no fellowship, be it of ever so learned or royal a society, no knightship,—not though it be of the Garter,—confers so fair an honour.”
—Anthony Trollope (1815–1882)