Table 2.

Evaluation measures

Measure	Formula	Evaluation focus
Accuracy	$\frac{TP+TN}{TP+FP+FN+TN}$	Degree of correctness (ability to classify a data point correctly)
Precision	$\frac{\mathit{TP}}{TP+FP}$	Proportion of positive identifications was actually correct
Recall	$\frac{\mathit{TP}}{TP+FN}$	Proportion of actual positives was identified correctly
F-measure	$\frac{2TP}{2TP+FP+FN}$	Combination of precision and recall
AUC	$\frac{1}{2}\left(\frac{\mathit{TP}}{TP+FN}+\frac{\mathit{TN}}{TN+FP}\right)$	Degree of separability (the ability to distinguish between classes)

TP true positive, TN true negative, FP false positive, FN false negative