/* * BioJava development code * * This code may be freely distributed and modified under the * terms of the GNU Lesser General Public Licence. This should * be distributed with the code. If you do not have a copy, * see: * * http://www.gnu.org/copyleft/lesser.html * * Copyright for this code is held jointly by the individual * authors. These should be listed in @author doc comments. * * For more information on the BioJava project and its aims, * or to join the biojava-l mailing list, visit the home page * at: * * http://www.biojava.org/ * */ /* * @(#)NormalizingKernel.java 0.1 00/01/20 * * By Thomas Down */ package org.biojava.stats.svm; /** * Performs a normalization on the results of a nested kernel. *

* This is equivalent to making the locations in feature space of the nested * kernel unit vectors lying on a unit sphere. The dot product in feature space * then becomes just cos theta rather than * ||a|| * ||b|| * cos theta as both lengths are 1. The length of * a in the feature space of kernel k is sqrt( k(a, a) ), so that * the normalizing kernel ends up calculating * k(a, b) / sqrt( k(a, a) * k(b, b) ). *

* As the values of k(x, x) are required repeatedly, it may be worth making the * nested kernel a DiagonalCachingKernel. * * @author Thomas Down * @author Matthew Pocock */ public class NormalizingKernel extends NestedKernel { public NormalizingKernel() {} public NormalizingKernel(SVMKernel k) { setNestedKernel(k); } public double evaluate(Object a, Object b) { SVMKernel k = getNestedKernel(); double kAA = k.evaluate(a, a); double kBB = k.evaluate(b, b); double kAB = k.evaluate(a, b); return kAB / Math.sqrt(kAA * kBB); } public String toString() { return "Normalizing Kernel K(x, y | k) = " + " k(x, y) / sqrt(k(x, x) * k(y, y)); k(x,y) = " + getNestedKernel().toString(); } }