ziadatalabs/FreeSyntheticInsuranceClaims100M

Free Synthetic Insurance Claims (100M)

A synthetic dataset of 100,000,000 property & casualty insurance claims — auto, home, renters, property, and business lines. Built for fraud-detection modeling, claims-triage pipelines, and general insurance-analytics prototyping.

Scope note: this dataset covers property & casualty claims only. There is no health-insurance or clinical content of any kind — no diagnoses, treatments, or medical data anywhere in the schema.

No real policyholders, claims, or insurer data were used — every record is generated from scratch.

Schema

Column	Type	Description
claim_id	string	Unique identifier for the claim
policy_id	string	Policy the claim was filed against (policies can have multiple claims)
claim_type	string	auto, home, renters, property, or business
state	string	US state where the policy is held
policyholder_tenure_years	float	Years the policyholder has held the policy
previous_claims_count	int	Number of prior claims on the policy
incident_date	date	Date the incident occurred
claim_filed_date	date	Date the claim was filed (always on or after incident_date)
claim_amount	float	Dollar amount claimed
deductible	float	Policy deductible applied to the claim
claim_status	string	approved, denied, pending, or under_review
days_to_resolution	float	Days from filing to resolution (null while pending)
is_fraud_flagged_ground_truth	bool	Ground-truth fraud flag

Format

Single Parquet file, Snappy-compressed.

Quick start

import pandas as pd
df = pd.read_parquet("insurance_claims_100M.parquet")
print(df.head())

# Or with duckdb for larger-than-memory queries
import duckdb
duckdb.sql("SELECT claim_type, AVG(claim_amount) FROM 'insurance_claims_100M.parquet' GROUP BY claim_type")

# Or with the datasets library
from datasets import load_dataset
ds = load_dataset("ziadatalabs/FreeSyntheticInsuranceClaims100M")

Notes

• The fraud ground-truth label follows a realistic underlying pattern rather than being random: claims are far more likely to be flagged when the policyholder has held the policy for less than a year and the claim amount is unusually high relative to other claims of the same type — a well-documented real-world red flag (new-policy-plus-outsized-claim), not an arbitrary label.
• Claim status is correlated with the fraud flag (flagged claims are denied or sent to review far more often than approved outright), so a status-prediction model has a genuine signal to learn from.
• days_to_resolution is only populated for resolved/in-progress claims; it's null for every pending row.
• This dataset is part of a growing collection of free synthetic datasets across security, finance, healthcare operations, retail, geospatial, gaming, and other domains.

License & Usage

Licensed under CC BY-NC 4.0 (Creative Commons Attribution-NonCommercial 4.0). Free to use for personal, research, and educational purposes with attribution. Not licensed for commercial use.

---

Published by Zia Data Labs. More free synthetic datasets at huggingface.co/ziadatalabs.

Want more free datasets? Hit the ❤️ and follow. And we take requests — tell us what synthetic data you need, and we'll build it.

Contact: zia.data.team@protonmail.com