X is a matrix of size 64400 and each column

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: wever, there is no overlap for the two classes and they are seperated pretty. Thus, FDA is better than PCA here. Practical e xample of 2_3 In this matlab example we explore FDA using our familiar data set 2_3 which consists of 200 handwritten "2" and 200 handwritten "3". X is a matrix of size 64*400 and each column represents an 8*8 image of "2" or "3". Here X1 gets all "2" and X2 gets all "3". >la 23 >od _ >X =X: 120; >1 (, :0) >X =X: 2140; >2 (, 0:0) Next we calculate within class covariance and between class covariance as before. >m1=ma(1 2; >u enX, ) >m2=ma(2 2; >u enX, ) >s =(u -m2 *(u -m2' >b m1 u) m1 u); >s =cvX' +cvX'; >w o(1) o(2) We use the first two eigenvectors to project the dato in a two- dimensional space. >[ d =eg(ivs)*s ) >v ] is n(w b; >w=v: 12; > (, :) >Xht=w*; >_a 'X Finally we plot the data and visualize the effect of FDA. > satroe(,0)Xht120) > cte(ns120,_a(:0) > hl o > od n > satroe(,0)Xht2140,r) > cte(ns120,_a(0:0)'' wikicour senote.com/w/index.php?title= Stat841&pr intable= yes 21/74 10/09/2013 Stat841 - Wiki Cour se Notes FDA projection of data 2_ 3, using Matlab (http://www.mathwork.com) . Map the data into a linear line, and the two classes are seperated perfectly here. An e xte ns ion of Fis he r's dis criminant analys is for s tochas tic proce s s e s A general notion of Fisher's linear discriminant analysis can extend the classical multivariate concept to situations that allow for function- valued random elements. The development uses a bijective mapping that connects a second order process to the reproducing kernel Hilbert space generated by its within class covariance kernel. This approach provides a seamless transition between Fisher's original development and infinite dimensional settings that lends itself well to computation via smoothing and regularization. Link for Algorithm introduction:[[11] (http://statgen.ncsu.edu/icsa2007/talks/HyejinShin.pdf) ] FD...
View Full Document

This document was uploaded on 03/07/2014.

Ask a homework question - tutors are online