A cave-bat-count model that misses an actual roost night is the expensive error. The matrix is 15 true roosts called roost, 5 missed roosts, 4 false roost calls, and 76 true quiet nights called quiet. What is recall (sensitivity), and what does it mean?
Select an answer to reveal the explanation.
Short Explanation
TP 15, FN 5. Recall is 15 / 20 times 100 percent, which is 75 percent: confidence that actual positives were not missed. 79 percent is precision. Missed roosts are the expensive error, so recall is the named metric.
Full Explanation
TP = 15, FN = 5, so recall = 15 / 20 × 100% = 75%. It is confidence that actual positives were found. 79% is precision (15/19). 91% is accuracy ((15+76)/100). 25% inverts the ratio. Missed roosts are the costly miss, so recall is the named metric.