False alarms are very costly for a model that flags accounts. Which metric matters most?
Why C is the answer
Precision
Precision measures how many flagged cases were truly positive.
In detail
- Precision measures how many flags are correct.
- Precision looks at flagged cases.
Use case
An illustrative scenario, not a real incident.
Freezing a customer's account wrongly is costly, so the team optimizes for precision.
See it in different ways
flowchart TD
A{"Which error hurts most?"} -->|"False alarms"| B["Precision"]
A -->|"Missed cases"| C["Recall"]
A -->|"Both, with imbalanced classes"| D["F1 score"]
A -->|"Numeric predictions"| E["RMSE"]mindmap
root((Metrics))
Classification
Recall: catch every case
Precision: avoid false alarms
Regression
RMSE
Text
ROUGEflowchart LR S0["Model flags cases"] --> S1["Count true positives"] --> S2["Divide by all flagged"]
Key point: Precision counts how many flags were right.
Why the other options are wrong
- A. Training loss alone
- Loss does not measure false alarms.
- B. Number of epochs
- Epochs are not a metric.
- D. Recall alone
- Recall alone ignores false alarms.