Catching Cyber-Fraud

Example: A large credit card provider losing millions from cyber-fraud needs multiple data sources to improve under-performing fraud detection models.


  • Datasets contain significantly fewer examples of fraudulent transactions, skewing data and undermining model performance.
  • Training data from competitors,payment processors, and banks is needed, but trade secrets and privacy concerns impede access.
  • Model training thus slows without the necessary data, giving cyber criminals sufficient time to develop new strategies before existing models can improve outcomes.


  • Devron allows data scientists to train cyber-fraud models on third-party sets without directly accessing the data.
  • Diversified data sources would minimize skew and mean that the company’s models were better prepared to identify edge cases and sophisticated schemes.
  • Recurring model updates incorporate fresh data and could allow cyber-fraud models to stay one step ahead in identifying new methods of abuse.
Catching Cyber-Fraud
Catching Cyber-Fraud
< View all Use Cases and Solutions

See how Devron can provide better insight for your organization

Request a Demo