R&D / Banking AI Evaluation Library / Datasets and sources
PaySim synthetic mobile money transactions
Synthetic transactions from the PaySim mobile money simulator, with a fraud flag on each transaction.
Dodone links to the publisher and does not host or copy this data. Read the licence on the publisher page before you use it.
About this source
- Publisher
- Edgar Lopez-Rojas
- Context
- Simulated mobile money transactions
- Version
- Not available (Kaggle shows the last update as about ten years ago)
- Last checked
- 2026-11-03
- Labels or answers
- A fraud flag for each transaction, from the simulator.
Facts stated by the publisher
- The Kaggle page describes the data as made by the PaySim simulator.
Licence, access and privacy
- Licence
- Creative Commons Attribution-ShareAlike 4.0 International (opens in a new tab) (Licence verified)
- Scope
- The Kaggle page states this licence. Changed versions that are shared must use the same licence.
- Access
- Registration needed. Downloads from Kaggle need a Kaggle account.
- Privacy
- Privacy reviewed: low risk. Fully synthetic data from a simulator.
Limitations
- A fraud flag does not say why a payment failed or what the right next action is.
- Mobile money patterns differ from bank payment operations.
- The data has not been updated for many years.
Suggested for these evaluations
- Payment exception triage Diagnostic only
Why it helps: Synthetic transactions can test whether an agent reads transaction patterns correctly.
What it does not validate: It has no payment exceptions, causes or approved remedies, so it cannot judge a full triage. - Financial-crime alert analysis Diagnostic only
Why it helps: Fraud flags on synthetic transactions test a narrower pattern-finding skill.
What it does not validate: Money-laundering patterns and investigation outcomes.