Given the training data below, we want to train a binary classifier using Naive Bayes. The last column is the class label y, and each column of x denotes a categorical feature. Features Sky, Humid, and Wind have two possible values; feature Temp can take three values (cold, mild, hot). Problem 5 continued on next page. . . 2 ,_ - .. .... i1 1').. _ 'd)_ input features a: = (5.91, .132, 033,034) Class label y Sky Temp Humid Wind Play Sport sunny mild normal strong yes rainy cold high mild no rainy hot normal strong no sunny hot high mild no sunny cold normal mild yes sunny mild normal strong no rainy mild high strong no rainy mild normal mild yes sunny hot normal strong yes sunny cold high strong no a) How many independent parameters are there in your Naive Bayes classifier? Iustify your answer. (5pt) b} What are the maximum likelihood estimates for these parameters? (10pt) c} Suppose we have a new input vector x = (sunny, cold, normal, strong). What is P(y = 1|x)? Which class label will the Naive Bayes classifier (with the parameters from part (b)) assign to this example? Iustify your answer. (Spt) d} (Bonus question) Suppose we have a new input vector x = (sunny, cold, normal, missing). That is, you are given values for the first three features only. What is P(y = 1|x) according to the Naive Bayes model from part (b)? Justify your answer. (Spt)