Framework Engineering Research

An open index of evidence, theory, and practice

research-document

False Negative Risk

Research Area
General Research
Status
draft
Version
0.1
Priority
medium
Updated
2026-07-22
Source
research/evidence-runs/ECR-000001-multi-layer-procedural-representation/comparison/normalizer-validation/false-negative-risk.md

False Negative Risk

False negatives occur when equivalent procedural concepts fail to normalize together.

High-risk cases:

  • wording variation
  • passive voice
  • domain-specific synonyms
  • generic wording
  • nested object values
  • source text vs value fields

Why this matters:

  • false negatives understate conceptual overlap
  • false negatives can make ontology gaps look like model disagreement