a) Consider the problem of predicting if a person has a college degree based on age and salary

a) Consider the problem of predicting if a person has a college degree based on age and salary. The table below contains training data for 10 individuals. Salary ($) College Degree 40,000 Yes 52,000 No 25,000 No 77,000 Yes 48,000 Yes 110,000 33,000 44,000 27,000 65,000 Build a decision tree for classifying whether a person has a college degree or not using classification errors. What is the depth of your tree and classification error? b. A multivariate decision tree is a generalization of univariate decision trees, where more than one attribute can be used in the decision rule for each split. That is, splits need not be orthogonal to a feature's axis. For the same data, learn a multivariate decision tree where each decision rule is a linear classifier that makes decisions based on the sign of axage + BXincome — 1. What is the depth of your tree, as well as a and B? c. Multivariate decision trees have practical advantages and disadvantages. List two advantages and two disadvantages multivariate decision trees have compared to univariate decision trees.