QuestionQ97

VerifiedML Functional Performance Metrics

After testing a model that detects rotten tomatoes, a test engineer observed the following data from a certain number of tomato images.

Question Image

For this confusion matrix, which set of values for accuracy, recall, and specificity, respectively, is correct?

Explanation

With rotten tomatoes as the positive class, there are 45 true positives, 5 false negatives, 42 true negatives, and 8 false positives. Thus, accuracy is (45 + 42) / 100 = 0.87, recall is 45 / (45 + 5) = 0.9, and specificity is 42 / (42 + 8) = 0.84.

Community Discussion

No comments yet. Be the first to start the discussion!