Slide 1
Slide 2
1. Evaluating Classifiers
1. Evaluating Classifiers(cont.)
2. Plain Accuracy and Its Problems
2. Plain Accuracy and Its Problems(cont.)
3. The Confusion Matrix (CM)
3. The Confusion Matrix (cont.)
3. The Confusion Matrix(cont.)
3. The Confusion Matrix (cont.)
3. The Confusion Matrix (cont.)
3. The Confusion Matrix (cont.)
3. The Confusion Matrix (cont.)
4. Problems with Unbalanced Classes
4. Problems with Unbalanced Classes (cont.)
4. Problems with Unbalanced Classes (cons.)
4. Problems with Unbalanced Classes (cons.)
5. Problems with Unequal Costs & Benefits
5. Problems with Unequal Costs & Benefits (cont.)
6. A Key Analytical Framework: Expected Value
6. A Key Analytical Framework: Expected Value(cont.)
6. A Key Analytical Framework: Expected Value (cont.)
6. A Key Analytical Framework: Expected Value (cont.)
7. Using Expected Value to Frame Classifier Use
7. Using Expected Value to Frame Classifier Use(cont.)
7. Using Expected Value to Frame Classifier Use(cont.)
7. Using Expected Value to Frame Classifier Use(cont.)
8. Using Expected Value to Frame Classifier Evaluation
8. Using Expected Value to Frame Classifier Evaluation(cont.)
8. Using Expected Value to Frame Classifier Evaluation(cont.)
8. Using Expected Value to Frame Classifier Evaluation(cont.)
8. Using Expected Value to Frame Classifier Evaluation(cont.)
8. Using Expected Value to Frame Classifier Evaluation(cont.)
8. Using Expected Value to Frame Classifier Evaluation(cont.)
8. Using Expected Value to Frame Classifier Evaluation(cont.)
Slide 36
10. Summary
Slide 38
Targeting the Best Prospects for a Charity Mailing
Targeting the Best Prospects for a Charity Mailing (cont.)
Slide 41
Slide 42
Slide 43
Slide 44
A Brief Digression on Selection Bias
A Brief Digression on Selection Bias(cont.)
Our Churn Example Revisited with Even More Sophistication
Slide 48
Slide 49
Slide 50
Slide 51
Slide 52
Assessing the Influence of the Incentive
Assessing the Influence of the Incentive (cont.)
Assessing the Influence of the Incentive (cont.)
From an Expected Value Decomposition to a Data Science Solution
Slide 57
Slide 58
Summary