The past paper question about Machine Learning: A research team is developing a

Question

The past paper question about Machine Learning:

A research team is developing a system to help screen for people at risk of developing a rare, but life- threatening medical condition. People identified by the system as "at risk" will be monitored for signs of the condition. The team are evaluating an experimental binary classifier using a test set, and have obtained the test results shown in the following confusion matrix What is the overall accuracy of the classifier on the test set? You should show your working. What is the true positive rate (recall) of the classifier. How effective is the classifier in identifying people who are "at risk"?

Explanation / Answer

i) overall accuracy of the classifier = (total actual at risk / total no of people screened) * 100 = (25/200) *100 = 12.5 %

ii) true positive rate of the classifier = (45/200) * 100 = 22.5 %

iii) the classifier is 12.5 % effective in identifying people who are at risk.

Related Questions

Navigate

• Browse (All)
• Browse T
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.