1(癌症) | 0(非癌症) | |
1(预测为癌症) | True Positive | False Positive |
0(预测为非癌症) | False Nagative | True Negative |
判断癌症病人的分类器好坏标准:
1.准确率(Precision) = 预测且实际得癌症数目 / 预测得癌症数目 = (TP/(TP + FP))
//P = 1-误报率(假阳)
2.召回率(Recall)= 预测且实际得癌症数目 / 实际得癌症数目 = (TP / (TP + FN))
//R=1-漏报率(假阴)
阈值 threshold 样本分类标准
if( threshold ==0.99) : high P ,low R (找的对)
else if( threshold == 0.01) :low P,high R (找的全)
3.F-Measure:F-Measure是Precision和Recall加权调和平均
F = (a*a +1)*P*R/(a×a ×P + R)
当a = 1时,
if((P == 0) || (R == 0)):F = 0
else if ((P = 1) && (R = 1)): F = 1