更新时间:2023-12-03 23:14:46
它实际上取决于您的特定应用程序. 假设您要尽量减少误报的数量(因为处理任何误报的后果将花费您很多钱)
It actually depends on your specific application. Say you want to minimize number of false positives (because it will cost you a lot to deal with consequences of any false alarm)
在这种情况下,选择第一个分类器,因为它的误报率小于第二个分类器的误报率:
In this case, choose the first classifier, because its false positive rate is less then that of the second classifier:
58/(58 + 124)< 61/(61 + 127) 0.3186813< 0.3244681
58/(58+124) < 61/(61+127) 0.3186813 < 0.3244681
在这里看看 http://en.wikipedia.org/wiki/Accuracy_and_precision
在这里: http://en.wikipedia.org/wiki/Sensitivity_and_specificity
如果您只想***的分类器"-您会遇到问题,因为两个分类器具有相同的准确性:
If you just want "the best classifier" - you have a problem, since both classifiers have the same accuracy:
a1 =(349 + 124)/(349 + 124 + 58 + 93)= 0.7580128 a2 =(346 + 127)/(346 + 127 + 61 + 90)= 0.7580128
a1 = (349+124)/(349+124+58+93) = 0.7580128 a2 = (346+127)/(346+127+61+90) = 0.7580128
因此,您需要分析您的领域或行业并决定是否要:
So you need to analyze your domain or industry and decide whether you want to:
1)尽可能减少误报-然后选择误报率最低的分类器;
1) get as little false alarms as possible - then choose classifier with minimum false positive rate;
2)尽可能少地漏掉案件-然后选择具有最低误报率的分类器;
2) get as little missed cases as possible - then choose classifier with minimum false negative rate;
3)争取更多点击率-然后选择具有最高真实阳性率的分类器;
3) get more hits as possible - then choose classifier with maximum true positive rate;
4)获得更正确的拒绝率-然后选择具有最大真实否定率的分类器.
4) get more correct rejections - then choose classifier with maximum true negative rate.