Tom is the Credit Manager at ABC Enterprises, a furniture and technology retailer, wants to develop a decision model to optimise the probability that good customers will be allowed to have credit accounts. while customers at risk of becoming a bad debt will not be allowed a credit account. The following table sets out actual credit purchases for the past two months. Credit is currently allowed at the discretion of the store manager on the day of the purchase. It is this decision that Tom wants to formalise with a decision model. {Fest credit perfonnence) Case no. Income Credit rating Credit stews {=worst, 1 III-best] 1 154 B Good 2 11D 3 Good 3 55 B Good 4 55 B Good 5 12K] 3 Bad 150 9 Good ? 35 4 Bad 3 f5 5 Good 9 B5 5 Bad 1 QC! 5 Bad 11 115 6 Good 1 2 11 5 7 Good 1 3 1 t) T Good 1 4 Q5 6 Bad 1 5 Qt] T Bad 1 E- 1 [it] B Good 1 i' ?5 3 Bad 1 8 Q5 8 Good 1 9 13b 9 Good on 1 es 0 end 154 Good 110 Good 55 Good 65 Good 120 Bad 150 Good 35 Bad 75 Good 85 Bad 90 Bad 110 Good 115 Good 100 Good 95 Bad 90 Bad 16 100 Good 17 75 Bad 18 95 Good 19 130 Good 20 125 Bad 21 85 Good 22 115 Good 23 125 Good 24 105 Bad 25 115 Good 26 90 Bad Part a) Plot the data and create a decision model to three levels. Part b) Calculate the weighted average Ginni impurity score for your model. Part c) Explain how the model helps Tom to formalise the credit application process at ABC Enterprises