True rate, false positive rate, false negative rate, true negative rate of confusion matrix

 

True Positive (true, TP) is predicted by the model as a positive positive sample;

True Negative (true negative, TN) is predicted as a negative negative sample by the model;

False Positive (false positive, FP) is predicted as a positive negative sample by the model;

False Negative (false negative, FN) is predicted as a negative positive sample by the model;

one)

True Positive Rate (TPR) or sensitivity (sensitivity) 
TPR = TP / (TP + FN) 
number of positive sample prediction results / actual number of positive samples 

two)

True Negative Rate (TNR) or specificity (specificity) 
TNR = TN / (TN + FP) 
number of negative sample prediction results / actual number of negative samples

three)

False Positive Rate (False Positive Rate, FPR) 
FPR = FP / (FP + TN) the 
number of negative sample results predicted to be positive / the actual number of negative samples

Four)

False Negative Rate (FNR) 
FNR = FN / (TP + FN) 
the number of positive sample results predicted to be negative / the actual number of positive samples

for example

 

 

reference

This article gives you a thorough understanding of accuracy, precision, recall, true rate, false positive rate, ROC/AUC - AIQ (6aiq.com)