Definition
The accuracy with which automated classification assigns reportable incident types and regulatory event categories, measured against expert re-classification on a stratified sample, with the false-negative direction reported separately.
The Measurement Gap
There is no metric that tells you how often a reportable fire, exposure, or clinical event is classified into a non-reportable category by automated classification, which means submission completeness scores well and the regulatory count is understated at the source.
Formula
RECA-FER_FN = M_reportable→non / R_true ; reported per category with the full confusion matrix
Proposed formula, pending validation pilots.
At a Glance
Decision Supported
Whether automated NFIRS and NEMSIS classification can be relied on for regulatory submission, and which categories require human classification.