Methods Based On Proximity Matrices
A method based on proximity matrices is one where the data is presented to the algorithm in the form of a similarity matrix or a distance matrix. These methods all fall under the broader class of metric multidimensional scaling. The variations tend to be differences in how the proximity data is computed; for example, Isomap, locally linear embeddings, maximum variance unfolding, and Sammon mapping (which is not in fact a mapping) are examples of metric multidimensional scaling methods.
Read more about this topic: Nonlinear Dimensionality Reduction
Famous quotes containing the words methods, based and/or proximity:
“I conceive that the leading characteristic of the nineteenth century has been the rapid growth of the scientific spirit, the consequent application of scientific methods of investigation to all the problems with which the human mind is occupied, and the correlative rejection of traditional beliefs which have proved their incompetence to bear such investigation.”
—Thomas Henry Huxley (1825–95)
“Whoever invented the meeting must have had Hollywood in mind. I think they should consider giving Oscars for meetings: Best Meeting of the Year, Best Supporting Meeting, Best Meeting Based on Material from Another Meeting.”
—William Goldman (b. 1931)
“Reckoned physiologically, everything ugly weakens and afflicts man. It recalls decay, danger, impotence; he actually suffers a loss of energy in its presence. The effect of the ugly can be measured with a dynamometer. Whenever man feels in any way depressed, he senses the proximity of something “ugly.” His feeling of power, his will to power, his courage, his pride—they decline with the ugly, they increase with the beautiful.”
—Friedrich Nietzsche (1844–1900)