Large Datasets
In practice, a large data set leads to a large K, and storing K may become a problem. One way to deal with this is to perform clustering on your large dataset, and populate the kernel with the means of those clusters. Since even this method may yield a relatively large K, it is common to compute only the top P eigenvalues and eigenvectors of K.
Read more about this topic: Kernel Principal Component Analysis
Famous quotes containing the word large:
“Roughly speaking, any man with energy and enthusiasm ought to be able to bring at least a dozen others round to his opinion in the course of a year no matter how absurd that opinion might be. We see every day in politics, in business, in social life, large masses of people brought to embrace the most revolutionary ideas, sometimes within a few days. It is all a question of getting hold of them in the right way and working on their weak points.”
—Aleister Crowley (1875–1947)