Answer Happy

The simple Naive Bayes model is given in the following. F F F) 2 The training dataset includes 1000 samples. The random variable Y represents the label, which is either class A, B, or C. We have three features represented by random variables F1, F2, and F3 All the three features are binary, i.e., (0.1). The count of samples for each feature labelled by each class is shown below. F F2 F: Total Label Y А B с Total 0 100 300 100 500 400 0 100 500 0 150 150 50 350 1 350 150 150 650 0 50 0 150 200 1 450 300 50 800 500 300 200 1000 1. What are the maximum likelihood estimates for the tables PLY). P(F1IY), and P(F2IY), and P(F3|Y)? Y PLY) А 00 B с
Label PIF /V) P(F2Y) Fi-O F-1 F2-0 F2-1 А B IC Label P(F31) Y F3=0 Fs=1 P B IC 2. Consider a new data point (F:-1,F2-1, F3-1). what label would the classifier assign to this sample? Anser
3. Let's use Lapalce Smoothing to smooth out our distribution. Compute the new distribution for P(F1) given Laplace Smoothing with k=100. Label P(F11) Y F2=0 F1=1 A A B с Note that the blanks are filled with a fraction or a number with 3 decimal points.