Construct and interpret a confusion matrix from review-queue results.
Instruction: Among 1,000 cases, 100 are truly positive. A model flags 80 cases, including 60 true positives. State the four confusion-matrix counts, accuracy, precision, recall and F1. Explain why accuracy alone is insufficient.
Context: Tests construction and interpretation of a binary confusion matrix from counts, including a rare-positive baseline.
Updated
Official answer available
Read the opening below, then unlock the full answer and practical guidance.
Define rows as actual class and columns as predicted class. There are 60 true positives, 20 false positives, 40 false negatives and 880 true negatives. These counts sum to 1,000; the 100 actual positives are 60 + 40, and the 80 flagged cases are 60 + 20.
Accuracy is (60 + 880)...