Reportable Classification Accuracy (RCA)

GMI-LE-045Law EnforcementIntegration-RequiredPattern R-2: NLP Classification

Definition

The accuracy with which automated classification assigns mandatory reporting attributes, including bias motivation, domestic violence relationship, and reportable use of force.

The Measurement Gap

There is no metric that audits automated assignment of mandatory reporting attributes against source narratives, which means submitted figures understate exactly the categories that draw the most scrutiny.

Formula

RCA per attribute = 1 - ( FN_a + FP_a ) / N_a, with FN reported separately because the failure is asymmetric.

Proposed formula, pending validation pilots.

At a Glance

Unit
Percentage
Trend Direction
Higher
Category
Oversight, Disclosure, and Regulatory Conformance

Decision Supported

Whether mandatory reporting figures submitted to state and federal programs can be relied on.

← Regulated Task Credential Verification (RTCV)Retention Purge Conformance Rate (RPCR) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment