This assignment asks you to examine the ensemble model for classification.Please follow the detailed guidelines below for preparing and submitting your work. Click the title link to upload and submit them.

The attached file contains information about 5,000 customer data of Universal Bank, a relatively young bank.(This dataset is provided by one of the books in this class, "Data Mining for Business Analytics.")

The bank is interested in growing customers to bring in more loan business. The bank encouraged the marketing department to come up with an idea for better target marketing and to determine what factors make a customer accept a personal loan. The attached dataset includes demographic variables of its customers, customer response to the personal loan campaign (Personal Loan), etc.

Question 1. Using the dataset build a decision tree using the CART methodology in RStudio. Train, test, and plot your decision tree. Provide a detailed explanation of the splitting criteria and rules created by the CART model. Assess and report the accuracy of your results. Explain the outputs from the CART model and its performance. (35 pts)

Question 2: In RStudio Develop ensemble models, including bagging, boosting, and random forests, using the same dataset. Compare the outcomes of these models with the CART model you developed earlier. Discuss whether any of the ensemble methods improve upon the CART model's output. Provide detailed explanations for the differences observed in the results. (35 pts)

Question 3: Using the profile of a potential customer named Carolinewho is 32 years old, has six years of experience, graduated from law school with an education level of 3, earns about $200k yearly, and has a family but no mortgage loans, along with multiple security accounts and credit cardspredict whether she would accept a personal loan.Feel free to determine the number of family members for Caroline. Choose the best performing model from your previous analyses (CART or one of the ensemble models). Justify your choice of model and interpret the prediction results. (30 pts)