The tool

Forge a Mobile Money fraud-signal dataset

Choose how many transaction records you want and a seed. The same two numbers will always produce the same dataset — write the seed down and cite it in your methodology chapter.

Capped at 20,000 — the API generates synchronously (see ROADMAP.md Phase 1 for the planned async path beyond that).

Optional — leave both blank for the default (30 days from 2024-01-01). Transactions are granular, sub-day events, so this line defaults to a much narrower window than an insurance line's policy-year span. Transaction timestamps are spread across this range; end must be after start.

Two-stage additionally computes loss given fraud for fraudulent transactions — the loss_given_fraud_ghs column only exists in this mode. Single-stage produces the fraud-probability columns only.

Advanced: customize the model

Same directional relationships described in the methodology — strengthen, weaken, or rebalance them here without touching any code. Defaults are illustrative, anchored against publicly-documented Mobile Money fraud typologies, not derived from real transaction data.

Fraud risk (frequency)

Every factor below feeds one logistic model over is_fraudulent. Raise or lower a loading to strengthen or weaken that hypothesis.

Loss given fraud (severity) two-stage only