-
Precision vs. Recall & AUC plotML 2022. 6. 11. 19:56
매일 헷갈리는 precision, recall 정리해본다.
예측/정답 O X Positive TP FP Negative FN TN Precision : TP/(TP+FP)
Recall : TP/(TP+FN)
FP : Type I error
FN : Type II error
AUC plot 의 y 축 = Sensitivity = True Positive Rate = Recall
AUC plot 의 x 축 = 1 - Specificity = False Positive Rate = FP / (FP+FN) = 1 - FN/ (FP+FN)
'ML' 카테고리의 다른 글
ALS vs. SGD (0) 2022.06.11 Entropy, Cross Entropy, KLD (0) 2022.06.11 Clustering Validation - Prediction Strength (0) 2022.06.10 Causal Inference - causal mediation analysis vs DID vs SCM vs causal impact (0) 2022.06.10 PCA vs. SVD vs. FA (0) 2022.06.09