Question 5 True/False with short explanations. State whether the following statements are true or false and provide brief explanations to support your answer. a) Under the Nave Bayes assumption, we also require that P(A, B)=P(A)x PB) b) The formula of total description length is given by the following: Cost(Model Data) = Cost(Data|Model) + ax Cost(Model) Cost(Data Model): Number of bits needed to encode the misclassified instances Cost(Model): Number of bits needed to encode the model. Given two competing models, the model with higher total description length is preferred. c) Smaller decision trees should always be preferred over larger trees as they have better generalization performance. d) The classification performance of decision trees will degrade in the presence of redundant attributes (attributes are duplicates of each other)