BBitMonkey AI
Home / Case studies / P07
P07Applied Prediction and Automation

Expanded Package

Fraud detection

Sign in to open this case study.

Sign in

The case

PayNimbus is a payment gateway that processed 79,991 card transactions last month. 799 of them were fraud - one in a hundred, high for the industry and normal for a young gateway with a risky merchant mix. The review team can open 1,200 alerts a month. The existing rules engine raises 5,968.

What is in it

  • 13 teaching steps, one animation each
  • 26 labs, checked live against the case data
  • 65 step questions, every option explained
  • 20 final paper questions, marked on submission

The steps

  1. 1Eighty thousand payments, 1,200 reviews
  2. 2Where the fraud label comes from
  3. 3What a 1% base rate does to every number you know 2 labs
  4. 4What you know in two hundred milliseconds
  5. 5The model that agrees with the rules engine
  6. 6The rules engine you are replacing
  7. 7Train on three weeks, test on the fourth
  8. 8How deep to work the queue
  9. 9What the analyst actually sees
  10. 10Week four: they changed
  11. 11Responding without over-reacting
  12. 12The monitoring pack
  13. 13Blocking somebody's card