Two bicycle-share “needs-service” models each score 100 bikes. Model A: 20 true positives, 15 false alarms, 5 missed bad bikes, 60 true negatives. Model B: 20 true positives, 5 false alarms, 15 missed bad bikes, 60 true negatives. Missed bad bikes staying in service is the costly error. Which model matches that cost story, and why?

Short Explanation

Model A piles up false alarms. Model B piles up missed bad bikes. Missed bad bikes staying in service is the costly error, so Model B matches that story: recall is about 57 percent versus Model A's 80 percent. Precision on B is 80 percent. The extra errors on B are false negatives.

Full Explanation

Model A’s extra errors are false alarms (precision 20/35 ≈ 57%, recall 80%). Model B’s extra errors are missed bad bikes (precision 80%, recall 20/35 ≈ 57%). The costly miss is FN, so low recall on B matches the story.

×