一、ROC，AUC

ROC（Receiver Operating Characteristic）曲線和AUC常被用來評價一個二值分類器（binary classifier）的優劣?。

ROC曲線一般的橫軸是FPR，縱軸是FPR。AUC為曲線下面的面積，作為評估指標，AUC值越大，說明模型越好。如下圖：

?

二、Precision、Recall、F1-score

Terminology and derivations
from a?confusion matrix

true positive (TP) eqv. with hit true negative (TN) eqv. with correct rejection false positive (FP) eqv. with?false alarm,?Type I error false negative (FN) eqv. with miss,?Type II error sensitivity?or?true positive rate (TPR) eqv. with?hit rate,?recall specificity?(SPC) or?true negative rate (TNR) precision?or?positive predictive value?(PPV) negative predictive value?(NPV) fall-out?or?false positive rate (FPR) false discovery rate?(FDR) miss rate or?false negative rate?(FNR) accuracy?(ACC) F1 score is the?harmonic mean?of?precision?and?sensitivity Matthews correlation coefficient?(MCC) Informedness = Sensitivity + Specificity - 1 Markedness = Precision + NPV - 1

中文解釋

轉載于:https://www.cnblogs.com/tectal/p/9420452.html

總結

以上是生活随笔為你收集整理的评估指标：ROC，AUC，Precision、Recall、F1-score的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。