FAR 和 FRR 用于表示生物识别设备的结果。下面是由 weka 生成的生物特征数据生成的混淆矩阵。我找不到任何资源来解释使用 *n 混淆矩阵计算 FAR 和 FRR 的过程。任何解释程序的帮助都会有很大帮助。提前致谢!
Weka 还给出了这些值,TP Rate、FP Rate、Precision、Recall、F-Measure 和 ROC Area。请建议是否可以使用这些计算所需的值。
=== 混淆矩阵 ===
a b c d e f g h i j k l m n o <-- classified as
1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 | a = user1
0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 | b = user2
0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 | c = user3
0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 | d = user4
0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 | e = user5
0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 | f = user6
0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 | g = user7
0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 | h = user9
1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 | i = user10
0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 | j = user11
0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 | k = user14
0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 | l = user15
0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 | m = user16
0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 | n = user17
0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 | o = user19