混淆矩阵

Reference

一般用混淆矩阵来看分类器的表现，概念相对简单.

基本数据：

1. true positives (TP): These are cases in which we predicted yes (they have the disease), and they do have the disease.
2. true negatives (TN): We predicted no, and they don't have the disease.
3. false positives (FP): We predicted yes, but they don't actually have the disease. (Also known as a "Type I error.")
4. false negatives (FN): We predicted no, but they actually do have the disease. (Also known as a "Type II error.")

指标:

1. 正确率 Accuracy:(TP+TN)/total
2. 错误率 Misclassification Rate:(FP+FN)/total
3. True Positive Rate: TP/actual yes
4. False Positive Rate: FP/actual no
5. Specificity: TN/actual no
6. Precision: TP/predicted yes
7. Prevalence: actual yes/total

Change log

• 2017.10.28 创建