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:
“With a generous endowment of motherhood provided by legislation, with all laws against voluntary motherhood and education in its methods repealed, with the feminist ideal of education accepted in home and school, and with all special barriers removed in every field of human activity, there is no reason why woman should not become almost a human thing. It will be time enough then to consider whether she has a soul.”
—Crystal Eastman (1881–1928)
“The fetish of the great university, of expensive colleges for young women, is too often simply a fetish. It is not based on a genuine desire for learning. Education today need not be sought at any great distance. It is largely compounded of two things, of a certain snobbishness on the part of parents, and of escape from home on the part of youth. And to those who must earn quickly it is often sheer waste of time. Very few colleges prepare their students for any special work.”
—Mary Roberts Rinehart (1876–1958)
“Our senses perceive no extreme. Too much sound deafens us; too much light dazzles us; too great distance or proximity hinders our view. Too great length and too great brevity of discourse tends to obscurity; too much truth is paralyzing.... In short, extremes are for us as though they were not, and we are not within their notice. They escape us, or we them.”
—Blaise Pascal (1623–1662)