Precision和Recall

来自wikipedia，https://en.wikipedia.org/wiki/Precision_and_recall

		True condition
	Total population	Condition positive	Condition negative	Prevalence = Σ Condition positive/Σ Total population	Accuracy (ACC) = Σ True positive + Σ True negative/Σ Total population
Predicted condition	Predicted condition positive	True positive	False positive, Type I error	Positive predictive value (PPV), Precision = Σ True positive/Σ Predicted condition positive	False discovery rate (FDR), probability of false alarm = Σ False positive/Σ Predicted condition positive
	Predicted condition negative	False negative, Type II error	True negative	False omission rate (FOR) = Σ False negative/Σ Predicted condition negative	Negative predictive value (NPV) = Σ True negative/Σ Predicted condition negative
		True positive rate (TPR), Recall, Sensitivity, probability of detection = Σ True positive/Σ Condition positive	False positive rate (FPR), Fall-out= Σ False positive/Σ Condition negative	Positive likelihood ratio (LR+) = TPR/FPR	Diagnostic odds ratio (DOR) = LR+/LR−	F1 score = 2/1/Recall + 1/Precision
		False negative rate (FNR), Miss rate = Σ False negative/Σ Condition positive	True negative rate (TNR), Specificity (SPC) = Σ True negative/Σ Condition negative	Negative likelihood ratio (LR−) = FNR/TNR

PRECISION=TP/TP+FP 查准率

RECALL=TP/TP+FN 查全率

假设有样本：正例90，反例10

比较三个不同的分类器

A:

TP86  FP2

FN4 TN8

ACC = 0.94

P = 86/88 =  0.98

R = 86/90 = 0.96

F1 = 0.97

B:

TP90  FP10

FN0 TN0

ACC = 0.9

P = 90/100

R = 90/90

F1 = 0.95

C:

TP90 FP0

FN0 TN10

ACC = 1

P= 1

R = 1

F1= 1