Calibration
Loading public event evidence
The validation register is opening the read-only event view.
Source files are never used as a runtime fallback.
Loading public event evidence
Source files are never used as a runtime fallback.
MODEL VALIDATION / ANALYST TRIAGE DEMONSTRATION
Full row-level event queries, fixed holdout evidence, and constrained review capacity.
Measured 1% analyst-review queue only. Not production-ready.
CONTROL 01 / PUBLIC EVENT QUERY
Query every allowlisted event row. Results are read-only, ordered by event time, and limited to 100 rows per request.
| Event time, UTC | Event ID | Merchant / category | Amount | Fraud label | Source |
|---|---|---|---|---|---|
| Jan 1, 2019, 6:30 PM | 1751 | fraud_Schroeder Grouphealth_fitness | $89.90 | NO | fraudTrain |
| Jan 1, 2019, 6:30 PM | 1752 | fraud_Volkman-Predovicshopping_net | $8.81 | NO | fraudTrain |
| Jan 1, 2019, 6:30 PM | 1753 | fraud_Gaylord-Powlowskihome | $14.71 | NO | fraudTrain |
| Jan 1, 2019, 6:31 PM | 1754 | fraud_Powlowski-Weimannfood_dining | $30.00 | NO | fraudTrain |
| Jan 1, 2019, 6:31 PM | 1755 | fraud_Romaguera and Sonstravel | $6.23 | NO | fraudTrain |
| Jan 1, 2019, 6:32 PM | 1756 | fraud_Baumbach, Hodkiewicz and Walshshopping_pos | $362.86 | NO | fraudTrain |
| Jan 1, 2019, 6:32 PM | 1757 |
CONTROL 02 / CAPACITY
Change the planning rate to see workload. Recall remains visible only for the measured 1% policy because no other threshold sweep is available.
EVIDENCE 03 / MODEL COMPARISON
Both policies train on the earlier simulated partition and evaluate on the later one. The challenger only adds class weighting to the same logistic model family. It loses on every reported measure and is not promoted.
Calibration
CONTROL 04 / DATA BOUNDARY
The complete simulated source is published through allowlisted analytical fields. Identity-like raw columns and decisioning capabilities remain outside the public view.
Not available here: identity-like raw columns, personal identifiers, model scores, scoring endpoints, payment decisions, or payment-processing behavior.
EVIDENCE 05 / COVERAGE LEDGER
These source totals reconcile the public event view to all 1,852,394 ingested rows.
| $43.90 |
| NO |
| fraudTrain |
| Jan 1, 2019, 6:32 PM | 1758 | fraud_Wiza LLCmisc_pos | $28.03 | NO | fraudTrain |
| Jan 1, 2019, 6:32 PM | 1759 | fraud_Deckow-Darefood_dining | $112.61 | NO | fraudTrain |
| Jan 1, 2019, 6:33 PM | 1760 | fraud_Larkin Ltdkids_pets | $48.81 | NO | fraudTrain |
| Jan 1, 2019, 6:36 PM | 1761 | fraud_Fadel, Mertz and Rippinentertainment | $124.42 | NO | fraudTrain |
| Jan 1, 2019, 6:37 PM | 1762 | fraud_Lowe, Dietrich and Erdmankids_pets | $220.47 | NO | fraudTrain |
| Jan 1, 2019, 6:37 PM | 1763 | fraud_Maggio-Faheymisc_pos | $3.15 | NO | fraudTrain |
| Jan 1, 2019, 6:37 PM | 1764 | fraud_Armstrong, Walter and Gottliebfood_dining | $82.57 | NO | fraudTrain |
| Jan 1, 2019, 6:39 PM | 1765 | fraud_Huel Ltdmisc_net | $100.93 | NO | fraudTrain |
| Jan 1, 2019, 6:39 PM | 1766 | fraud_Gleason-Macejkovicshopping_net | $47.79 | NO | fraudTrain |
| Jan 1, 2019, 6:40 PM | 1767 | fraud_Flatley Groupmisc_pos | $6.77 | NO | fraudTrain |
| Jan 1, 2019, 6:40 PM | 1768 | fraud_Thompson-Gleasonhealth_fitness | $3.97 | NO | fraudTrain |
| Jan 1, 2019, 6:40 PM | 1769 | fraud_Windler, Goodwin and Kovacekhome | $19.04 | NO | fraudTrain |
| Jan 1, 2019, 6:42 PM | 1770 | fraud_Hermann and Sonsshopping_pos | $3.63 | NO | fraudTrain |
| Jan 1, 2019, 6:42 PM | 1771 | fraud_Schmeler-Howepersonal_care | $26.54 | NO | fraudTrain |
| Jan 1, 2019, 6:43 PM | 1772 | fraud_Zboncak Ltdfood_dining | $7.08 | NO | fraudTrain |
| Jan 1, 2019, 6:45 PM | 1773 | fraud_Bradtke, Torp and Bahringerpersonal_care | $1.68 | NO | fraudTrain |
| Jan 1, 2019, 6:45 PM | 1774 | fraud_Pouros-Haagshopping_pos | $4.39 | NO | fraudTrain |
| Jan 1, 2019, 6:45 PM | 1775 | fraud_Metz-Boehmshopping_pos | $3.89 | NO | fraudTrain |
The dashed diagonal is perfect calibration. The observed curve is uneven, so scores should not be interpreted as reliable probabilities without further calibration work.
Review capacity
Review capacity stays fixed. Monthly volume changes with the holdout distribution. This is retrospective workload evidence, not a forecast.
Limits: simulated source; one chronological split; allowlisted amount, merchant, category, and prior-history inputs; no confidence intervals, fairness analysis, threshold sweep, live scoring, or evidence for automated payment decisions.