Binary Classification metrics
The significance level α
- the probability of making the wrong decision when the null hypothesis is true.
- Alpha levels (sometimes just called “significance levels”) are used in hypothesis tests.
- Usually, these tests are run with an alpha level of .05 (5%)
- but other levels commonly used are .01 and .10.
| Term | Formula | |
|---|---|---|
| Precision | PPV, Positive predictive value | True positive/Test outcome positive |
| Accuracy | (true positive + true negative)/Toral Population | |
| Sensitivity | True positive rate | True positive/Condition positive |
| Specificity | True negative rate | True negative/Condition negative |
| Prevalence | Condition positive/Total population | |
Precision(PPV, Positive predictive value)
Precision = True positive/Test outcome positive
Accuracy (ACC):
Accuracy = (true positive + true negative)/Toral Population
Precision-Recall (PR) curves
- recall (sensitivity, True positive rate) = true positives / (true positives + false negatives)(condition positive)
ROC curves and the AUC
