Financial Transaction Fraud Score

This dataset provides granular, real-time financial transaction records enriched with device, geo-location, and behavioral features, along with model-generated fraud risk scores. It is ideal for training, evaluating, and deploying machine learning models for fraud detection, and supports operational monitoring and regulatory reporting. The comprehensive schema enables deep analysis of transaction patterns and risk factors across accounts, devices, and geographies.

  • last updated 20 Aug 2025
  • by GoMask
The brief that made it

Real-time fraud detection and prevention in banking and payments

Sample rows

preview · 8 of 200 rows · all 23 columns
transaction_idstringtransaction_typestringfraud_scorefloatis_fraudbooleanmerchant_categorystringaccount_idstringtransaction_datetimedatetimetransaction_amountfloatcurrencystringmerchant_idstringdevice_idstringdevice_typestringdevice_osstringip_addressstringgeo_latitudefloatgeo_longitudefloatlocation_citystringlocation_statestringlocation_countrystringis_internationalbooleanprevious_transaction_count_24hintegerprevious_transaction_amount_24hfloatmodel_versionstring
TXN000001purchase0.18falsegroceryACC10012024-06-01T09:15:23Z24.99USDMCH001DEV10001mobileiOS192.168.0.4540.7128-74.006New YorkNYUSfalse382.45v2.3.0
TXN000002purchase0.42falsetravelACC10022024-06-01T11:23:51Z2500GBPMCH002DEV10002desktopWindows51.140.34.20051.5074-0.1278LondonblankGBtrue23200v2.4.1
TXN000003payment0.09falseonline servicesACC10032024-06-02T14:02:19Z0.99USDMCH003DEV10003tabletAndroid172.16.254.134.0522-118.2437Los AngelesCAUSfalse715.21v2.3.0
TXN000004transfer0.86trueblankACC10042024-06-02T16:40:55Z10500EURblankDEV10004desktopLinux193.134.25.6748.85662.3522ParisblankFRtrue110800v2.3.0
TXN000005purchase0.12falsecoffee shopACC10052024-06-03T18:05:12Z9.49USDMCH004DEV10005mobileAndroid10.0.0.2541.8781-87.6298ChicagoILUSfalse528.76v2.4.1
TXN000006purchase0.27falseelectronicsACC10062024-06-03T20:47:33Z420.55AUDMCH005DEV10006desktopMacOS203.23.11.85-33.8688151.2093SydneyNSWAUtrue2415v2.5.1
TXN000007payment0.04falsecharityACC10012024-06-04T08:10:50Z0.01SGDMCH006DEV10007mobileiOS118.200.21.2201.3521103.8198SingaporeblankSGtrue20.01v2.3.0
TXN000008transfer0.92trueblankACC10072024-06-04T09:35:14Z14500BRLblankDEV10008desktopWindows186.201.15.50-23.5505-46.6333Sao PauloblankBRtrue318000v2.3.0

What the 200 rows show

from the 200-row sample

Transfer (transaction type) stands out: 26 of its 26 rows have is_fraud = true, against 2 of 174 for the rest.

  • 14%is_fraud = true
  • 0.17median fraud_score
  • 5device types
  • 5model versions
  • 7device oses
  • 11currencies
Is fraud rate by transaction_typeis_fraud = true
0%50%100%0%purcha…0 of 1070%withdr…0 of 3100%transf…26 of 266%payment2 of 360%refund0 of 28
fraud_score200 rows, in bands of 0.1
035706648281011428111200.51fraud_score →

Median 0.17, from 0.01 to 0.99.

merchant_category171 rows with a value · 29 left blank
  1. charity33
  2. coffee shop23
  3. electronics18
  4. grocery17
  5. education15
  6. fast food13
  7. online services11
  8. home improvement9
  9. luxury goods9
  10. automotive9
23 columns by typefrom the column list below
  • string 14
  • integer 1
  • float 5
  • datetime 1
  • boolean 2

Columns

23 columns in four groups
blueprint · 23 columns
columntypedescriptionexample
Text 14 columns
transaction_idstringUnique identifier for each financial transactionuniqueTXN000001
account_idstringUnique identifier for the account involved in the transactionACC1001
currencystringISO 4217 currency code for the transaction (e.g., USD, EUR)11 currenciesUSD
merchant_idstringUnique identifier for the merchant where the transaction occurredoptionalMCH001
merchant_categorystringCategory of the merchant (e.g., grocery, electronics, travel)12 categories · optionalgrocery
transaction_typestringType of transaction (e.g., purchase, withdrawal, transfer)purchase · withdrawal · transfer · payment · refundpurchase
device_idstringUnique identifier for the device used in the transactionoptionalDEV10001
device_typestringType of device used (e.g., mobile, desktop, tablet, POS terminal)mobile · desktop · tablet · POS terminal · other · optionalmobile
device_osstringOperating system of the device (e.g., iOS, Android, Windows)7 oses · optionaliOS
ip_addressstringIP address from which the transaction was initiatedoptional192.168.0.45
location_citystringCity where the transaction occurredoptionalNew York
location_statestringState or province where the transaction occurredoptionalNY
location_countrystringCountry where the transaction occurred (ISO 3166-1 alpha-2 code)11 countries · optionalUS
model_versionstringVersion of the fraud detection model used to generate the fraud score5 versions · optionalv2.3.0
Numbers 6 columns
transaction_amountfloatMonetary value of the transaction0 or more24.99
geo_latitudefloatLatitude of the transaction location-90 to 90 · optional40.7128
geo_longitudefloatLongitude of the transaction location-180 to 180 · optional-74.006
previous_transaction_count_24hintegerNumber of transactions by this account in the previous 24 hours0 or more · optional3
previous_transaction_amount_24hfloatTotal amount transacted by this account in the previous 24 hours0 or more · optional82.45
fraud_scorefloatFraud risk score assigned to the transaction (0.0 = low risk, 1.0 = high risk)0 to 10.18
Dates and times 1 column
transaction_datetimedatetimeTimestamp when the transaction occurred2024-06-01T09:15:23Z
True or false 2 columns
is_internationalbooleanIndicates if the transaction is international (true) or domestic (false)optionalfalse
is_fraudbooleanIndicates whether the transaction is confirmed as fraudulent (true) or not (false)optionalfalse

Use it for

  • is fraud14%28 of 200 rowsmean fraud score by t…0.22purc…0.64with…0.89tran…0.16paym…

    A finance dashboard

    The is_fraud rate, fraud_score by transaction_type and a breakdown of merchant_category. Excel, Power BI or Tableau.

  • Why do 28 of 200 rows have is_fraud = true?

    A root-cause class exercise

    Hand out the rows and one question. The answer is in the data, not in the brief.

  • A software demo

    Believable transactions with account_id, transaction_datetime and transaction_amount to fill a screen in front of a buyer.

Not quite right?

Make it yours.

Same 23 columns, your size and your rules. See 20 rows before you pay.

Preview 20 rows free

1,000 rows: $1.99 · 10,000: $12.99One-time. No subscription. All prices

This dataset200 rows23 columns
Yours1,000 rows23 columnsip_address: UK only

blueprint · financial-transaction-fraud-score

Behind this dataset

Same schema. As many rows as you need.

These 200 rows came out of a blueprint — 23 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
  • Transaction metadata with timestamps
  • Device ID and location tracked
  • Fraud scores generated via algorithms
  • Label for actual fraud cases
Rows
Open the blueprint in Data Factory

1 credit per row. Preview 20 rows free. Buy rows for the full run.

Exports
CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence
yours to use, including commercially
API slug
financial-transaction-fraud-score

More in Finance

Browse all
Also in FinanceCredit Card Fraud DetectionFinancial Statement Fraud IndicatorsBank Customer Segmentation DatasetAll 1,967 datasets

What should your data show?

Preview 20 rows free
No signup. No card.