Question 1 a) information gain is based on the decrease in entropy after splitting a dataset based on an attribute. the meaning of constructing a decision tree is all about finding the attributes having the........ information gain Choose from : minimum or maximum or equally -svm tried to choose the hyperplane which separates the data points as widely as possible since the margin .......... improve the model accuracy on the test or the unseen data. choose from minimization or maximization or has no impact to -suppose you ran logistic regression with regularization lambda =1 and once lambda = 0 which value of theta corresponds to lambda =0 choose from a) theta = 1.37 ,0.51 b) theta =74.81,45.05 b) -increasing the batch size can be used to solve the exploding gradient problem. true or false -When using feature scaling it speeds up the gradient descent by making it requires fewer iterations to get to the good solution. True or false