即使没有100％的准确度，为什么我仍得到1.000 ROC面积值 [英] Why am I getting a 1.000 ROC area value even when I don't have 100% of accuracy

问题描述

我正在使用Weka作为分类器，到目前为止，它对我来说非常有用。但是，在上一次测试中，我得到了1.000 ROC区域值（如果我没记错的话，它代表了一个完美的分类），而没有100％的准确性，如图中的混淆矩阵所示。

I am using Weka as a classifier, and it has worked great for me so far. However, in my last test, I got a 1.000 ROC area value (which, if i remember correctly, represents a perfect classification) without having 100% of accuracy, as can be seen in the Confusion Matrix in the Figure.

我的问题是：是我对结果的解释不正确还是得到错误的结果（也许我正在使用的分类器编程不正确，尽管我认为这不太可能）？

My question is: Am I interpreting the results incorrectly or am I getting wrong results (maybe the classifier I am using is badly programmed, although I don't think it's likely)?

分类输出

谢谢！

推荐答案

精度是在一个特定的阈值（通常为0.5）下测量的。
如果AUC为1，则意味着您还有另一个具有完美分类的阈值，对于您而言，我想应该是一个较低的阈值。

The accuracy is measured at one specific threshold, typically 0.5. If the AUC is 1, it means that you have an other threshold with perfect classification, in your case I would guess a lower one.

这篇关于即使没有100％的准确度，为什么我仍得到1.000 ROC面积值的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！