Question: Create and parameterize the naive Bayes model for the German credit scoring problem. Compute the following evaluation measures for both the default classifier (i.e., predicting
Create and parameterize the naive Bayes model for the German credit scoring problem.
Compute the following evaluation measures for both the default classifier (i.e., predicting all loans will be repaid, based on the prior probability of a 70% chance of repayment) and your NB model:
1. Predictive accuracy 2. AUC 3. Expected cost, using either (or both) of the cost tables discussed in the text 4. IRB
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
