Evaluating Standards

主要罗列一些机器学习中常用的评估方法

Recall And Precision

	Ground Truth Pos	Ground Truth Neg
Predict As Pos	True Pos	False Pos
Predict As Neg	False Neg	True Neg

$Precision = \frac{True Pos}{True Pos + False Pos} \\ Recall \hspace{8mm} = \frac{True Pos}{True Pos + False Neg} \\ F1_{score} (or F_{score}) = 2*\frac{precision*recall}{precision+recall}$