Question

Consider a 2-class pattern recognition problem with two dimensional feature vector x e R. The class conditional density for class-wų is uniform over [1,3]X[1,3] and for class W2 is uniform over [2,4]>[2,4]. The prior probability of two classes are as p(W1) = 0.5 and p(w2) = 0.5. (For this problem, assume that we are using 0-1 loss function and hence Bayes classifier minimizes probability of miss classification)
(a) Which is Bayes classifier? Is it unique for this problem?
(b) What would be the probability of misclassification of bayes classifier​

Answer 1

Explanation:

. All the patterns to recognize belong to J different classes, j=1,2 ... probability density function depends on the ... Example of classification using class- conditional probability