Exhibit 9-1. The confusion matrix for a classification method with Class 1 and C
ID: 3266901 • Letter: E
Question
Exhibit 9-1. The confusion matrix for a classification method with Class 1 and Class 0 is given below
Confusion Matrix
Predicted Class
Actual Class
1
0
1
120
10
0
100
70
Refer to Exhibit 9-1. What is the percent overall error rate?
36.67%
43.25%
58.82%
7.69%
Refer to Exhibit 9-1. What are the percent error rates for Class 1 and Class 0 ?
7.69% and 58.82%, respectively
36.67% and 43.25%, respectively
7.69% and 36.67% , respectively
43.25% and 7.69%, respectively
Confusion Matrix
Predicted Class
Actual Class
1
0
1
120
10
0
100
70
Explanation / Answer
Here,
Confusion Matrix
Predicted Class
Actual Class
1
0
1
a11=120
a10=10
0
a01=100
a00=70
Here, a11=120=is number of observations whose class labels are 1 & also whose predicted class labels are 1(correct classification) ,
a10= 10=is number of observations whose class labels are 1 but whose predicted class labels are 0(mis- classification) ,
a01=100=is number of observations whose class labels are 0 but whose predicted class labels are 1(mis- classification) ,
a00=70=is number of observations whose class labels are 0 & also whose predicted class labels are 0 (correct classification)
(1) The percent overall error rate is=(Number of mis-classified observations)/Total observations
={(a10+a01)/(a11+a10+a01+a00)}*100
={(10+100)/(120+10+100+70)*100}*100
=36.66667%
Hence , answer of this question is option A.36.67%
(2)The percent error rates for Class 1 ={a10/(a11+a10)}*100
={10/(120+10)}*100
=7.692308%
The percent error rates for Class 0 ={a01/(a01+a00)}*100
={100/(100+70)}*100
=58.82353%
Hence , answer of this question is option A.7.69% and 58.82%, respectively
Confusion Matrix
Predicted Class
Actual Class
1
0
1
a11=120
a10=10
0
a01=100
a00=70
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.