# Naive bayes classifier - digit recognition

• Published on
11-Feb-2015

• View
493

4

Embed Size (px)

DESCRIPTION

Navie Bayes Classifer home work digit recognition Global school of Media, Soongsil Univ. 09th, May, 2013 Heedeok Lee

Transcript

• 1. Naive Bayes classifier, 200914359TH, MAY, 2013
• 2.
• 3. 0 9 , .Ex) 2(10) 10(2) , .2 2 2
• 4.
• 5.
• 6. .) Helvetica Helvetica0 9 (10 )20 1 2 3 4 5 6 7 8 9
• 7. , , , , , , -
• 8. 1. > , 2. > 1 , 3. > ? ?4.
• 9. , ? THE MNIST DATABASE of handwritten digits 6 0 9 .http://yann.lecun.com/exdb/mnist/ 28x28 ,
• 10. Naive Bayes classifier . ? , 0 9 -> 0.1 ? -> ( ) ?1) 0 2) 1
• 11. Navie Bayes classifier - likelihood0,2 on = 0,2 on / = 2 / 3 = 0.6666
• 12. Naive Bayes classifier 3 ,{F0,0 = 0, F0,1 = 0, F0,2 = 1, ,F8,8=0} = x 0,0 8,8 0.9
• 13. Naive Bayes classifier1 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 0.10 0.11 0.012 0.053 0.054 0.305 0.806 0.907 0.058 0.609 0.500 0.801 0.052 0.013 0.904 0.805 0.906 0.907 0.058 0.609 0.500 0.80 3,1 5,5
• 14. Naive Bayes classifier2 3 = 0.1 = 0.12,2 On0.82,2 On0.95,4 On0.15,4 On0.86,4 Off0.16,4 Off0.80.0008 0.0567<
• 15. Naive Bayes classifier 2 3 = 0.1 = 0.12,2 On0.82,2 On0.95,4 On0.15,4 On0.86,4 Off0.16,4 Off0.80.00008 0>8,8 On0.018,8 On0
• 16. Pseudo codeInt bayesclassifier(input[][], likelihood[][][]){Int result[10];For(I = 0; I < 10; i++){Result[i] = 0.1;For(j = 0; j < 28; j++)For(k=0; k