A home improvement retail store is offering its customers store-branded credit cards that come with a deep discount when used to purchase in-store home improvement products. To maintain the profitability of this marketing campaign, the store manager would like to make these offers only to the customers who are likely to carry a high monthly balance on the credit card. A data set obtained from a nationwide association of home improvement stores contains records of 500 consumers who carry similar credit cards offered by other home improvement stores. Relevant variables include Sex (Female or Male), Education (1 if did not finish college, 2 if undergraduate degree, 3 if graduate degree), Children (1 if have children, 0 otherwise), Age (1 if below 20 years old, 2 if 20–29 years, 3 if 30–39 years, 4 if 40–49 years, 5 if 50–59 years, 6 if 60 years and older). Balance is the target variable where one indicates the customer maintains a high monthly balance, and 0 otherwise. A portion of the CreditCard_Data worksheet is shown in the accompanying table.

a. Partition the data to develop a naïve Bayes classification model. Report the accuracy, sensitivity, specificity, and precision rates for the validation data set. 

b. Generate the ROC curve. What is the area under the ROC curve (or AUC value)? 

c. Interpret the performance measures and evaluate the effectiveness of the naïve Bayes model. 

d. Score the new customer records in the CreditCard_Score worksheet. What is the scoring result of the first customer record?

e. Change the cutoff value to 0.3. Report the accuracy, sensitivity, specificity, and precision rates for the validation data set.  

Transcribed Image Text:

Sex Female Male I Female Education 2 1 ⠀ 2 Children Yes No ⠀ Yes Age 3 4 E 3 لیا Balance 1 0 ⠀ 0