ART-93 · Wave 19 · Sanctions · Fuzzy-Match Calibration

Fuzzy-Match Calibration Scorer

Given an engine config (algorithm, threshold) and a synthetic labelled name-pair set, computes FPR/recall/F1 and recommends threshold calibration. No real names — synthetic fixtures only. Part of the sanctions-fuzzy-calibration chain.

F1 · Recall · FPR Threshold Calibration · Wolfsberg Synthetic Pairs Only

🔒 All inputs are processed locally in your browser. No data is transmitted. Do not enter real personal data — use synthetic or anonymised inputs only.

Engine Configuration

FieldAlgorithm

FieldThreshold (0.00 – 1.00)

Synthetic Name Pairs

JSON ArrayName pairs — [{"name_a":"...","name_b":"...","is_match":true}]

Invalid JSON — check syntax and try again.

–

F1 Score

–

FPR

–

Recall

–

Calibration Grade

Confusion matrix

–

TP — True Positives

–

FP — False Positives

–

FN — False Negatives

–

TN — True Negatives

Threshold recommendation

–

RECOMMENDED THRESHOLD

Pairs evaluated

Compliance flags

References (verify current): Wolfsberg Sanctions Screening Guidance (2019) — scoring/calibration benchmarks · FATF Guidance on Sanctions-Related Due Diligence (2021). SYNTHETIC NAME PAIRS ONLY — outputs are decision-support drafts, not regulatory advice.

OpenChainGraph v0.4 artifact · execution_hash:

–