一、机器学习中算法常用的评测标准
二、Precision(查准率) and Recall(查全率)
以预测病人是否有癌症为例子,假设 y = 1 y=1 y=1代表病人有癌症。下面是我从吴恩达老师的机器学习课程中截取的图片:
True Positive(真正,TP): 将正类预测为正类数
True Negative(真负,TN): 将负类预测为负类数
False Positive(假正,FP): 将负类预测为正类数→误报
False Negative(假负,FN): 将正类预测为负类数→漏报
precision = true positives no. of predicted positive = TP TP+FP recall = true positives no. of actual positive = TP TP+FN \begin{aligned} & \text { precision } = \frac { \text { true positives } } { \text { no. of predicted positive } } = \frac{\text{TP}}{\text{TP+FP}} \\ & \text { recall } = \frac { \text { true positives } } { \text { no. of actual positive } }= \frac{\text{TP}}{\text{TP+FN}} \end{aligned}