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:
“The reading public is intellectually adolescent at best, and it is obvious that what is called “significant literature” will only be sold to this public by exactly the same methods as are used to sell it toothpaste, cathartics and automobiles.”
—Raymond Chandler (1888–1959)
“Our children evaluate themselves based on the opinions we have of them. When we use harsh words, biting comments, and a sarcastic tone of voice, we plant the seeds of self-doubt in their developing minds.... Children who receive a steady diet of these types of messages end up feeling powerless, inadequate, and unimportant. They start to believe that they are bad, and that they can never do enough.”
—Stephanie Martson (20th century)
“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)