E-Commerce AI Fraud Detection Logs

This dataset provides detailed logs from AI-powered fraud detection systems in online retail, including transaction context, user information, model outputs, and operational responses. It is designed to help managers and security teams benchmark fraud detection workflows, analyze response effectiveness, and monitor regulatory compliance. The comprehensive schema supports advanced analytics, process optimization, and audit readiness.

  • last updated 11 Jan 2026
  • by GoMask
The brief that made it

Benchmarking and improving AI fraud detection workflows

Sample rows

preview · 8 of 120 rows · all 21 columns
event_idstringresponse_actionstringresponse_time_secondsfloatcompliance_flagbooleanshipping_address_statestringevent_timestampdatetimetransaction_idstringuser_idstringuser_emailstringtransaction_amountfloatcurrencystringpayment_methodstringshipping_address_streetstringshipping_address_citystringshipping_address_postal_codestringshipping_address_countrystringfraud_scorefloatfraud_labelstringai_model_versionstringresponse_timestampdatetimenotesstring
EVT-1A9F8B6Capproved3trueIL2022-01-15T09:22:31ZTX-4F7B3C8AUSR-83920147john.smith23@gmail.com79.99USDcredit_card5025 West AveChicago60616US0.1legitimatev2.4.12022-01-15T09:22:34ZRoutine purchase; loyalty member
EVT-4C7E2D9Bapproved2trueblank2023-05-21T13:09:10ZTX-AB47X9W2USR-24470123ashley.green@outlook.com58.99EURpaypalRue de la Paix 18Paris75002FR0.2legitimatev2.4.32023-05-21T13:09:12ZRoutine purchase
EVT-AC8D309Emanual_review140trueShanghai2024-03-23T18:32:41ZTX-7D9E2B5CUSR-90281731li.wei.88@163.com215CNYbank_transferNo. 99 Jiangsu RdShanghai200050CN0.47suspiciousv2.5.0-beta2024-03-23T18:35:01ZHigh velocity; repeat user
EVT-92FB1E73BAmanual_review150trueblank2024-05-14T15:11:09ZTX-QW8DF2V9USR-17834001paul.turner@icloud.com320.5GBPdebit_card221B Baker StLondonNW1 6XEGB0.44suspiciousv2.5.12024-05-14T15:13:39ZUnusual location; flagged for review
EVT-7F3ABCD9rejected600falseMH2023-07-28T11:29:20ZTX-9K8L0P7SUSR-20938475milind.joshi@rediffmail.com1250INRcryptoPlot 42, MG RoadMumbai400001IN0.98fraudulentv2.5.22023-07-28T11:39:20ZZero value crypto; delayed response
EVT-2B8F7D1Capproved3trueblank2022-09-13T08:45:34ZTX-5E6M1V2NUSR-14320856elena.martinez@live.com34.5EURcredit_cardCalle Mayor 10Madrid28013ES0.22legitimatev2.4.52022-09-13T08:45:37ZRoutine purchase
EVT-0E4B6F8Amanual_review180trueblank2023-03-10T16:19:40ZTX-8N1Q2W3OUSR-72483019sophie.tanaka@yahoo.co.jp410.45JPYdebit_card2-4-1 ShibuyaTokyo150-0002JP0.36suspiciousv2.4.82023-03-10T16:22:40ZRepeat user
EVT-6A1C0D2F3Bapproved0trueblank2021-12-31T23:59:59ZTX-2L8S7H3JUSR-82104732blank0USDotherblankblankblankblank0legitimatev2.5.02021-12-31T23:59:59ZBoundary transaction; edge: zero transaction

What the 120 rows show

from the 120-row sample

Escalated (response action) stands out: mean response_time_seconds is 644.0, against 76.0 for the rest.

  • 71%compliance_flag = true
  • 16median response_time_seconds
  • 3fraud labels
  • 6payment methods
  • 10currencies
  • 19shipping address countries
Mean response_time_seconds by response_action120 rows
04008002.3approv…59 rows102.8manual…26 rows364.6reject…13 rows61.7flagged7 rows644.0escala…15 rows
response_time_seconds120 rows, in bands of 100
0459083140100220400700response_time_seconds →

Median 16, from 0 to 660.

shipping_address_state62 rows with a value · 58 left blank
  1. CA8
  2. IL6
  3. Shanghai5
  4. ON5
  5. MH3
  6. NSW3
  7. SP3
  8. NY3
  9. WA3
  10. Beijing3
21 columns by typefrom the column list below
  • string 15
  • float 3
  • datetime 2
  • boolean 1

Columns

21 columns in four groups
blueprint · 21 columns
columntypedescriptionexample
Text 15 columns
event_idstringUnique identifier for each fraud detection event log entryuniqueEVT-1A9F8B6C
transaction_idstringUnique identifier for the associated e-commerce transactionTX-4F7B3C8A
user_idstringUnique identifier for the user involved in the transactionUSR-83920147
user_emailstringEmail address of the user involved in the transactionoptionaljohn.smith23@gmail.com
currencystringISO 4217 currency code for the transaction (e.g., USD, EUR)8 valuesUSD
payment_methodstringPayment method used for the transaction6 valuescredit_card
shipping_address_streetstringStreet address for shippingoptional5025 West Ave
shipping_address_citystringCity for shipping addressoptionalChicago
shipping_address_statestringState or province for shipping addressoptionalIL
shipping_address_postal_codestringPostal code for shipping addressoptional60616
shipping_address_countrystringCountry for shipping addressoptionalUS
fraud_labelstringAI system's classification of the transaction (e.g., legitimate, suspicious, fraudulent)legitimate · suspicious · fraudulentlegitimate
ai_model_versionstringVersion identifier of the AI model used for fraud detectionv2.4.1
response_actionstringAction taken in response to the fraud detection eventapproved · manual_review · rejected · flagged · escalatedapproved
notesstringAdditional notes or comments about the event or responseoptionalRoutine purchase
Numbers 3 columns
transaction_amountfloatTotal monetary value of the transaction0 or more79.99
fraud_scorefloatAI-generated score indicating likelihood of fraud (0.0 to 1.0)0 to 10.1
response_time_secondsfloatTime taken (in seconds) from event detection to response action0 or more · optional3
Dates and times 2 columns
event_timestampdatetimeDate and time when the fraud detection event occurred2022-01-15T09:22:31Z
response_timestampdatetimeDate and time when the response action was executedoptional2022-01-15T09:22:34Z
True or false 1 column
compliance_flagbooleanIndicates if the event and response met regulatory compliance requirementstrue

Use it for

  • compliance flag71%85 of 120 rowsmean response time se…2.3appr…102.8manu…364.6reje…61.7flag…

    An e-commerce dashboard

    The compliance_flag rate, response_time_seconds by response_action and a breakdown of shipping_address_state. Excel, Power BI or Tableau.

  • Why do the 15 escalated rows have a mean response_time_seconds of 644.0?

    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 events with event_timestamp, transaction_id and user_id to fill a screen in front of a buyer.

Not quite right?

Make it yours.

Same 21 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 dataset120 rows21 columns
Yours1,000 rows21 columnsshipping_address_street: UK only

blueprint · e-commerce-ai-fraud-detection-logs

Behind this dataset

Same schema. As many rows as you need.

These 120 rows came out of a blueprint — 21 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 row represents a unique fraud event flagged by the AI system.
  • Include event timestamp, transaction amount, risk score, action taken, and fraud outcome (confirmed/false positive).
  • Flagged events may trigger one of several actions: auto-reject, manual review, or customer notification.
  • Outcomes must balance between confirmed fraud (20-35%), false positives (10-25%), and legitimate transactions (40-60%) to reflect operational realities.
  • Record the AI model version and response latency for each event to facilitate performance analysis.
  • Events may be associated with device, payment method, and user account risk profiles.
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
e-commerce-ai-fraud-detection-logs

More in E-commerce

Browse all
Also in E-commerceE-commerce Ethical Product TraceabilityCustomer Product Review DatasetProduct Recommendation DatasetAll 1,954 datasets

What should your data show?

Preview 20 rows free
No signup. No card.