Insurance
27 columns
200 rows
8 formats

Insurance Claims Fraud Detection

This synthetic insurance claims dataset provides detailed records of individual claims, including customer demographics, policy details, incident descriptions, and a fraud label for supervised learning. Designed for fraud detection and claim triage automation, the dataset enables advanced analytics and machine learning model development for the insurance industry.

opened 13 times
last updated 21 Jul 2025
by GoMask

The brief that made it

Training and benchmarking insurance fraud detection models

Sample rows

preview · 8 of 200 rows · all 27 columns

claim_idstring	claim_statusstring	customer_ageinteger	fraud_reportedboolean	customer_occupationstring	policy_idstring	customer_idstring	claim_datedate	incident_datedate	claim_amountfloat	claim_typestring	incident_typestring	incident_location_streetstring	incident_location_citystring	incident_location_statestring	incident_location_postal_codestring	incident_location_countrystring	customer_genderstring	customer_incomefloat	policy_start_datedate	policy_end_datedate	policy_coverage_typestring	number_of_claimantsinteger	police_report_availableboolean	witness_presentboolean	claim_filed_channelstring	incident_descriptionstring
CLM-00001	approved	32	false	software engineer	POL-02001	CUS-10001	2023-02-18	2023-02-14	18500.75	auto	collision	412 Maple Ave	Dallas	TX	75201	USA	female	72000	2021-01-01	2026-01-01	comprehensive	1	true	false	online	Rear-ended by distracted driver on highway.
CLM-00002	in_review	44	false	marketing manager	POL-02002	CUS-10002	2022-06-10	2022-06-05	345000	property	fire	1592 Oak Street	Toronto	ON	M4B 1B3	Canada	male	95000	2017-03-15	2027-03-15	premium	2	true	true	agent	Kitchen fire caused by unattended stove.
CLM-00003	submitted	23	false	student	POL-02003	CUS-10003	2021-12-01	2021-11-30	1200	health	prescription	90 College Rd	Boston	MA	02115	USA	female	7500	2021-01-01	2024-01-01	basic	1	blank	blank	online	Covered prescription medication required after surgery.
CLM-00004	in_review	29	false	consultant	POL-02004	CUS-10004	2023-08-15	2023-08-12	59000.5	auto	theft	715 16th Avenue	Vancouver	BC	V5K 0A1	Canada	male	68000	2022-01-01	2027-01-01	comprehensive	1	true	false	phone	Vehicle stolen from parking garage overnight.
CLM-00005	approved	52	false	business owner	POL-02005	CUS-10005	2024-01-12	2024-01-05	560000	property	flood	4801 Grand Blvd	Chicago	IL	60611	USA	female	120000	2020-06-01	2030-06-01	premium	3	false	true	agent	Basement flooded after severe rainstorm.
CLM-00006	approved	34	false	teacher	POL-02006	CUS-10006	2022-05-03	2022-04-30	4000.99	health	accident	25 River St	Austin	TX	78701	USA	male	51000	2019-09-15	2024-09-15	comprehensive	1	blank	blank	phone	Broken arm due to fall while cycling.
CLM-00007	approved	74	false	retired	POL-02007	CUS-10007	2023-03-22	2023-03-20	310000	life	death	3 Kings Road	Manchester	ENG	M2 4WU	UK	female	32000	2003-05-01	2025-05-01	premium	3	false	false	agent	Policyholder deceased after heart failure.
CLM-00008	submitted	26	false	chef	POL-02008	CUS-10008	2022-09-07	2022-09-05	9100	auto	vandalism	77 Elm Street	San Jose	CA	95112	USA	male	29000	2022-01-01	2025-01-01	third-party	1	false	false	phone	Car windows smashed in parking lot overnight.
CLM-00009	in_review	61	false	artist	POL-02009	CUS-10009	2023-07-15	2023-07-10	80000	property	theft	2122 Willow Lane	Calgary	AB	T2P 1J9	Canada	female	41000	2018-07-01	2028-07-01	comprehensive	2	true	false	agent	Jewelry stolen during home break-in.
CLM-00010	submitted	38	false	nurse	POL-02010	CUS-10010	2023-05-28	2023-05-25	3500.99	health	illness	19 Broadway Ave	Montréal	QC	H2X 3P4	Canada	other	48000	2020-03-01	2025-03-01	basic	1	blank	blank	online	Medical expenses for flu-related hospitalization.
CLM-00011	approved	50	true	real estate agent	POL-02011	CUS-10011	2021-11-12	2021-11-11	160000	property	flood	8018 5th Ave	Seattle	WA	98109	USA	female	97000	2016-06-01	2026-06-01	premium	2	false	true	agent	Flooded basement after water pipe burst.
CLM-00012	in_review	36	false	bartender	POL-02012	CUS-10012	2022-03-19	2022-03-15	25999.89	auto	collision	605 Main St	Philadelphia	PA	19103	USA	female	26000	2021-01-01	2024-01-01	basic	1	true	false	in_person	Minor collision at busy intersection, no injuries.
CLM-00013	closed	65	false	business owner	POL-02013	CUS-10013	2024-01-04	2023-12-31	900000	life	death	14 Rue Grimaldi	Monaco	MC	98000	Monaco	male	1200000	1995-01-01	2035-01-01	premium	4	false	false	agent	Policyholder deceased in natural disaster.
CLM-00014	submitted	19	false	student	POL-02014	CUS-10014	2022-10-30	2022-10-29	4750.65	health	illness	10 College Ave	Ottawa	ON	K1P 1J1	Canada	unknown	7000	2022-01-01	2025-01-01	basic	1	blank	blank	online	Emergency room visit for food poisoning.
CLM-00015	approved	40	false	delivery driver	POL-02015	CUS-10015	2023-06-10	2023-06-08	20300	auto	windshield_damage	223 Pine St	Denver	CO	80202	USA	other	31000	2022-06-01	2025-06-01	third-party	1	true	false	in_person	Windshield cracked by flying debris on highway.
CLM-00016	rejected	59	true	contractor	POL-02016	CUS-10016	2022-11-11	2022-11-01	89000	property	fire	202 Lakeview Dr	Houston	TX	77002	USA	male	42000	2019-03-15	2029-03-15	premium	2	false	true	phone	Fire started in garage, significant damage.
CLM-00017	submitted	21	false	student	POL-02017	CUS-10017	2023-04-13	2023-04-12	2200	health	accident	88 Campus Ln	Ann Arbor	MI	48109	USA	female	9500	2022-01-01	2025-01-01	basic	1	blank	blank	online	Sprained ankle during sports activity.
CLM-00018	in_review	28	false	bartender	POL-02018	CUS-10018	2022-07-18	2022-07-10	44000	auto	collision	17 Oakwood Rd	New York	NY	10001	USA	male	27000	2022-01-01	2026-01-01	comprehensive	2	true	true	online	Multi-car accident on freeway, moderate injuries.
CLM-00019	approved	27	false	marketing manager	POL-02019	CUS-10019	2022-08-20	2022-08-19	14999.99	auto	theft	2307 Main St	Los Angeles	CA	90015	USA	female	65000	2020-01-01	2025-01-01	comprehensive	1	true	false	online	Car stolen while parked at shopping mall.
CLM-00020	submitted	54	false	investor	POL-02020	CUS-10020	2022-04-16	2022-04-12	73250	property	flood	2267 Royal Rd	London	ENG	SW1A 1AA	UK	male	870000	2017-01-01	2027-01-01	premium	2	false	false	agent	Flooding in lower level due to blocked drainage.

What the 200 rows show

from the 200-row sample

Closed (claim status) stands out: mean customer_age is 66.2, against 37.8 for the rest.

10%fraud_reported = true
36median customer_age
4incident location countries
4customer genders
4policy coverage types
5claim types

Mean customer_age by claim_status200 rows

customer_age200 rows, in bands of 10

Median 36, from 18 to 84.

customer_occupation200 rows · top 10 of 36 values

student32
consultant17
retired16
bartender15
business owner14
teacher11
contractor11
chef10
software engineer9
delivery driver9

27 columns by typefrom the column list below

string 16
integer 2
float 2
date 4
boolean 3

Columns

27 columns in four groups

blueprint · 27 columns
column	type	description	example
Text 16 columns
`claim_id`	string	Unique identifier for each insurance claimunique	CLM-00001
`policy_id`	string	Unique identifier for the insurance policy associated with the claim	POL-02001
`customer_id`	string	Unique identifier for the customer submitting the claim	CUS-10001
`claim_type`	string	Type of insurance claim (e.g., auto, property, health, life)auto · property · health · life · other	auto
`incident_type`	string	Type of incident leading to the claim (e.g., collision, theft, fire, illness, death, etc.)	collision
`incident_description`	string	Free-text description of the incident provided by the claimantoptional	Rear-ended by distracted …
`incident_location_street`	string	Street address where the incident occurredoptional	412 Maple Ave
`incident_location_city`	string	City where the incident occurredoptional	Dallas
`incident_location_state`	string	State or province where the incident occurredoptional	TX
`incident_location_postal_code`	string	Postal code of the incident locationoptional	75201
`incident_location_country`	string	Country where the incident occurred4 countries · optional	USA
`claim_status`	string	Current status of the claimsubmitted · in_review · approved · rejected · closed	approved
`customer_gender`	string	Gender of the customermale · female · other · unknown · optional	female
`customer_occupation`	string	Occupation of the customeroptional	software engineer
`policy_coverage_type`	string	Type of coverage provided by the policy (e.g., comprehensive, third-party, basic, premium)4 types · optional	comprehensive
`claim_filed_channel`	string	Channel through which the claim was filed (e.g., online, phone, agent, in-person)online · phone · agent · in_person · other · optional	online
Numbers 4 columns
`claim_amount`	float	Total amount claimed by the customer0 or more	18500.75
`customer_age`	integer	Age of the customer at the time of claim0 or more · optional	32
`customer_income`	float	Annual income of the customer0 or more · optional	72000
`number_of_claimants`	integer	Number of people involved in the claim (e.g., passengers, dependents)1 or more · optional	1
Dates and times 4 columns
`claim_date`	date	Date the claim was filed	2023-02-18
`incident_date`	date	Date the incident occurred	2023-02-14
`policy_start_date`	date	Date when the insurance policy startedoptional	2021-01-01
`policy_end_date`	date	Date when the insurance policy ends or expiredoptional	2026-01-01
True or false 3 columns
`fraud_reported`	boolean	Label indicating whether the claim is fraudulent (true) or not (false)	false
`police_report_available`	boolean	Indicates if a police report was filed for the incidentoptional	true
`witness_present`	boolean	Indicates if there were witnesses to the incidentoptional	false

Use it for

An insurance dashboard
The fraud_reported rate, customer_age by claim_status and a breakdown of customer_occupation. Excel, Power BI or Tableau.
Why do the 18 closed rows have a mean customer_age of 66.2?
A root-cause class exercise
Hand out the rows and one question. The answer is in the data, not in the brief.
Claims200
CLM-0000132approved
CLM-0001150approved
CLM-0001659rejected
A software demo
Believable claims with policy_id, customer_id and claim_date to fill a screen in front of a buyer.

This dataset200 rows27 columns

Yours10,000 rows27 columnsincident_location_street: UK only

blueprint · insurance-claims-fraud-detection

Behind this dataset

Same schema. As many rows as you need.

These 200 rows came out of a blueprint — 27 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.

Rules it was built with

Each claim assigned a fraud risk label
Include various claim types (auto, home, health)
Simulate realistic submission patterns and delays
Include customer, policy, and incident details
Claims may be flagged as 'investigated' or not

Rows1,000 rows10,000 rows100,000 rows1,000,000 rows

Open the blueprint in Data Factory

1 credit per row. New accounts start with 25 free credits.

Exports: CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence: yours to use, including commercially
API slug: insurance-claims-fraud-detection

What should your data show?

No signup. No card.

Sample rows

What the 200 rows show

Columns

Use it for

An insurance dashboard

A root-cause class exercise

A software demo

Make it yours.

Same schema. As many rows as you need.

More in Insurance

What should your data show?