6 1 2 Bayes V S Naive Bayes Consider Two Classes With The Same Priors P C1 P C2 The Same Covariance 0 Matrix 1 (52.55 KiB) Viewed 21 times
[6 1] = 2. Bayes v.s. Naive Bayes: Consider two classes with the same priors P(c1) = P(c2) the same covariance 0 matrix: Σ and means uli (1,1)" and Mi (2, 1)". Consider also the following test points: I=(0,1)T and y = (1.5,0)T. (a) How would x and y be classified according to full Bayes classifier? Show your computation. Note that the pdf of multivariate normal distributions is P(XC) N (Hi, with PDF: 1 (T – M;)";"(t – Mi)}. fi(2) exp{- (27)*|S| 2 109 Note also that V627)12:1 will be the same for both classes. The inverse of the covariance !-! (2πΙΣ | Σ. (b) Would the classification result differ if we use Naive Bayes? Explain why.
Join a community of subject matter experts. Register for FREE to view solutions, replies, and use search function. Request answer by replying!
with PDF: 1 (T – M;)";"(t – Mi)}. fi(2) exp{- (27)*|S| 2 109 Note also that V627)12:1 will be the same for both classes. The inverse of the covariance !-! (2πΙΣ | Σ. (b) Would the classification result differ if we use Naive Bayes? Explain why.