Unit 10 | Evaluating and Improving Models |
| 8.5 hrs |
Upon completion of this module, you will be able to:
- evaluate classification models with prediction probabilities
- measure classification performance with confusion matrices
- calculate Cohen's kappa coefficient to measure classification accuracy
- determine sensitivity, specificity, precision, and recall
- compare classification models via the F-measure
- plot Receiver Operating Characteristics (ROC) curves and determine AUC
- apply k-fold cross-validation for repeated holdout validation
- use bootstrap sampling
- improve ensemble learners with boosting
- train random forest models
- build ensemble models with stacking and bagging to improve prediction accuracy
Evaluating Model Performance
|
Required Work
Additional ResourcesSlide Deck & Data Sets
|
Improving Model Performance
|
Required Work
Additional Resources |