Detect Pig Butchering Victims
“Among accounts with escalating deposits to crypto exchanges over 90 days, which will make another large deposit in the next 14 days?”
Book a demo and get a free trial of the full platform: research agent, fine-tune capabilities, and forward-deployed engineer support.
By submitting, you accept the Terms and Privacy Policy.
Loved by data scientists, ML engineers & CXOs at
A real-world example
Among accounts with escalating deposits to crypto exchanges over 90 days, which will make another large deposit in the next 14 days?
$3.6B US losses in 2023. Victims groomed on dating apps are directed to fake crypto platforms. Signature: escalating deposits over weeks. Chen Zhi / Prince Group ($12B seized). A 15-minute welfare call costs $25; the average victim loses $150K. Banks have a duty-of-care obligation.
How KumoRFM solves this
Graph-powered fraud intelligence
Kumo’s backward window identifies the escalation pattern: A001 deposited $2K (Nov), $5K (Dec), $15K (Jan) — all to crypto exchanges. The model predicts another $5K+ deposit within 14 days with 87% probability. The relational graph reveals the pattern: increasing amounts, decreasing time between deposits, all flowing to the same exchange category.
From data to predictions
See the full pipeline in action
Connect your tables, write a PQL query, and get predictions with built-in explainability — all in minutes, not months.
Your data
The relational tables Kumo learns from
Accounts
| account_id | account_holder | account_type | age_bracket |
|---|---|---|---|
| A001 | Jane D. | Retail | 55-64 |
| A002 | Mark R. | Retail | 35-44 |
Transactions
| txn_id | account_id | amount | destination_type | timestamp |
|---|---|---|---|---|
| T001 | A001 | 2,000 | crypto_exchange | 2024-11-05 |
| T002 | A001 | 5,000 | crypto_exchange | 2024-12-10 |
| T003 | A001 | 15,000 | crypto_exchange | 2025-01-08 |
Write your PQL query
Describe what to predict in 2-3 lines — Kumo handles the rest
PREDICT SUM(TRANSACTIONS.AMOUNT WHERE TRANSACTIONS.DESTINATION_TYPE = 'crypto_exchange', 0, 14, days) > 5000 FOR EACH ACCOUNTS.ACCOUNT_ID WHERE SUM(TRANSACTIONS.AMOUNT WHERE TRANSACTIONS.DESTINATION_TYPE = 'crypto_exchange', -90, 0, days) > 3000
Prediction output
Every entity gets a score, updated continuously
| ACCOUNT_ID | TIMESTAMP | TARGET_PRED | True_PROB |
|---|---|---|---|
| A001 | 2025-02-01 | True | 0.87 |
| A002 | 2025-02-01 | True | 0.73 |
| A003 | 2025-02-01 | False | 0.05 |
Understand why
Every prediction includes feature attributions — no black boxes
Account A001 (Jane D.)
Predicted: 87% probability of next $5K+ crypto deposit
Top contributing features
Crypto exchange deposits (90d sum)
$22,000
39% attribution
Deposit escalation rate
2.5x per month
26% attribution
Destination type consistency
100% crypto_exchange
17% attribution
Account age_bracket
55-64
11% attribution
Account type
Retail
7% attribution
Feature attributions are computed automatically for every prediction. No separate tooling required. Learn more about Kumo explainability
PQL Documentation
Learn the Predictive Query Language — SQL-like syntax for defining any prediction task in 2-3 lines.
Python SDK
Integrate Kumo predictions into your pipelines. Train, evaluate, and deploy models programmatically.
Explainability Docs
Understand feature attributions, model evaluation metrics, and how to build trust with stakeholders.
Bottom line: Trigger customer welfare outreach on escalation patterns. A 15-minute call costs $25; average victim loses $150K. Banks have a duty-of-care obligation.
Related scenarios
Explore more fraud predictions
Topics covered
