uid stringclasses 10
values | dataset_id stringclasses 1
value | policy_type stringclasses 4
values | region stringclasses 3
values | underwriting_risk_score stringclasses 3
values | premium_level stringclasses 2
values | exposure_level stringclasses 3
values | early_claim_signal stringclasses 3
values | loss_ratio_trend stringclasses 2
values | underwriting_summary stringclasses 10
values | question stringclasses 1
value | ground_truth_label stringclasses 2
values | ground_truth_rationale stringclasses 10
values | outcome_signal stringclasses 6
values | source_citation stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
INSU-001 | insurance-underwriting-loss-coherence-risk-v0.1 | auto | UK | low | adequate | moderate | none | stable | Low risk profile aligned with claims. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | coherent | Premium and exposure aligned with outcomes. | stable_loss_ratio | Synthetic |
INSU-002 | insurance-underwriting-loss-coherence-risk-v0.1 | property | US | low | low | high | minor | upward | Underpriced high exposure property. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | incoherent | Premium insufficient for risk. | loss_ratio_rising | Synthetic |
INSU-003 | insurance-underwriting-loss-coherence-risk-v0.1 | health | EU | moderate | adequate | moderate | none | stable | Balanced underwriting. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | coherent | Risk and premium balanced. | stable | Synthetic |
INSU-004 | insurance-underwriting-loss-coherence-risk-v0.1 | commercial | US | high | adequate | high | major | upward | Large claims emerging. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | incoherent | Claims exceed expected risk. | loss_spike | Synthetic |
INSU-005 | insurance-underwriting-loss-coherence-risk-v0.1 | auto | UK | moderate | adequate | moderate | none | stable | Aligned pricing. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | coherent | Expected losses matched. | stable | Synthetic |
INSU-006 | insurance-underwriting-loss-coherence-risk-v0.1 | property | US | low | low | high | minor | upward | Storm risk underestimated. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | incoherent | Exposure mispriced. | loss_trend_up | Synthetic |
INSU-007 | insurance-underwriting-loss-coherence-risk-v0.1 | health | EU | low | adequate | low | none | stable | Low risk pool. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | coherent | Claims within model. | stable | Synthetic |
INSU-008 | insurance-underwriting-loss-coherence-risk-v0.1 | commercial | US | moderate | low | high | major | upward | High liability exposure underpriced. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | incoherent | Premium too low for exposure. | loss_ratio_spike | Synthetic |
INSU-009 | insurance-underwriting-loss-coherence-risk-v0.1 | auto | UK | moderate | adequate | moderate | none | stable | Risk priced correctly. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | coherent | Loss ratio steady. | stable | Synthetic |
INSU-010 | insurance-underwriting-loss-coherence-risk-v0.1 | property | EU | high | adequate | high | none | stable | High premium matches risk. | Is underwriting-loss coherence intact. Answer coherent or incoherent. | coherent | Premium covers risk. | stable | Synthetic |
What this repo is for
Detect misalignment between underwriting assumptions and real losses.
Focus
pricing vs exposure
risk score vs claim trend
early signals before loss spikes
Why it matters Insurers often discover pricing mistakes too late. This dataset tests whether systems can detect coherence loss early.
- Downloads last month
- 14