XTRACTIS FOR fraud detection

Detection of Fraudulent Credit Card Transactions

Benchmark vs. Logistic Regression, Random Forests, Boosted Trees & Neural Networks

Design, from only some transactions characteristics, an AI-based decision system that efficiently diagnoses credit card transactions, in order to rationally and instantly detect the fraudulent ones and to eventually adapt protection measures.

PROs & benefits

Identify the parameters involved in the fraudulent transactions and enhance knowledge by helping banking specialists understand the causal relationships between these parameters, their combination, and the occurrence of fraud (i.e., understand the scammers’ strategies).

Help the banking sector to make transparent decisions through automatic, explainable detection, while improving the consumer experience.

Use a detection model with fewer transaction characteristics to speed up protection process.


We get a Predictive Model that is:

A Decision System composed of 7 gradual rules without chaining, each using some of the 9 variables that XTRACTIS identified as significant (out of 29 potential predictors characterizing each transaction).
Very good Real Performance on External Test.
Efficient & Operational.
Running in real-time up to 70,000 predictions per second (i7 @2.5GHz with 8 physical cores), offline or online (API).
UC20 graphe benchmark

LoR=Logistic Regression | RFo=Random Forests | BT=Boosted Trees | NN=Neural Networks

Detailed results and explanations in full document.

Use Case 2023/11 (v2.0)

Powered by XTRACTIS® REVEAL 12.2.45294 (2023/06)


  1. Problem Definition
  2. XTRACTIS-induced Decision System
  3. XTRACTIS Process
  4. Top-Model Induction
  5. Explained Predictions for 2 unkown cases
  6. Top-Models Benchmark
  7. Appendices