Personal Loan Acceptance. Universal Bank is a relatively young bank growing rapidly in terms of overall customer acquisition. The majority of these customers are liability customers (depositors) with varying sizes of relationship with the bank. The customer base of asset customers (borrowers) is quite small, and the bank is interested in expanding this base rapidly to bring in more loan business. In particular, it wants to explore ways of converting its liability customers to personal loan customers (while retaining them as depositors).

A campaign that the bank ran last year for liability customers showed a healthy conversion rate of over 9% success. This has encouraged the retail marketing department to devise smarter campaigns with better target marketing. The goal is to use k-NN to predict whether a new customer will accept a loan offer. This will serve as the basis for the design of a new campaign.

The file UniversalBank.csv contains data on 5000 customers. The data include customer demographic information (age, income, etc.), the customer’s relationship with the bank (mortgage, securities account, etc.), and the customer response to the last personal loan campaign (Personal Loan). Among these 5000 customers, only 480 (=9.6%) accepted the personal loan that was offered to them in the earlier campaign.

Data Preprocessing. Transform the target attribute to Binominal type, and then specify the positive class (loan acceptance) as acceptor (replacing true) and the other class as nonacceptor (replacing false) using the Map operator in RapidMiner. Ensure that Education, a categorical predictor, is of Polynominal type. Also ensure that binary (0/1) predictors, namely, CD Account, CreditCard, Online, and Securities Account, are configured as Binominal type attributes. Consider all attributes as predictors except ID and ZIP Code.

Partition the data into training (60%) and holdout (40%) sets.

a. Consider the following customer: Age = 40, Experience = 10, Income = 84, Family = 2, CCAvg = 2, Education = 2, Mortgage = 0, Securities Account = false, CD Account = false, Online = true, and Credit Card = true. Perform a k-NN classification with the selected predictors using k = 1. Use the default threshold value of 0.5. How would this customer be classified?

b. What is a choice of k that balances between overfitting and ignoring the predictor information? Use RapidMiner’s Optimize Parameters (Grid) operator with nested 10-fold cross-validation on the training set to experiment with different k values, maximizing model accuracy. Report the confusion matrix, accuracy, precision, and recall for the 10-fold cross validation performance (averaged) for the best k.

c. Show the confusion matrix for the holdout data that results from using the best k. Report and interpret classification performance metrics for the holdout set.

d. Consider the following customer: Age = 40, Experience = 10, Income = 84, Family = 2, CCAvg = 2, Education = 2, Mortgage = 0, Securities Account = false, CD Account = false, Online = true, and Credit Card = true. Classify the customer using the best k.

e. Would you recommend using accuracy as the metric for finding the best k for this business context? Comment on whether any alternative performance metrics, if any, would be better for optimizing model performance in finding the best k.