1. Build a decision tree for this data set, using entropy as splitting criterion. Use two-way splits for the features "homeowner" and "income",as well as three-way splits for the feature "marital status".Do notconduct any splits of zero information gain, and stop once all remaining splitsgiveaninformationgainofzero.
2. What is the accuracy of your tree on the training data?Construct aconfusion matrix.
3. What does your tree predict (default yeso) for a new borrower whichdoesnotownahome,whichismarried andwhichhasahighincome? Howconfident is thisprediction?

HOME OWNER MARITAL STATUS INCOME DEFAULTED Bl no married lowr no B2 yes divorced high no B3 no single high yes B4 no married low no BS no married lowr yes B6 yes single high no B7 no married high no B8 no single low no B9 yes married high no B10 yes divorced high yes