Data & Innovation | Financial Services
Detection and correction of fraud using Fraud Detector
Fraud prevention using a machine learning model created by Nubiral: Fraud Detector.
About the client
An important company that breaks the traditional scheme of financial transactions in Mexico through technology and innovation.
It has been offering payment and collection solutions since 2008 in the following areas: electronic wallets, electronic vouchers, credit and debit cards on VISA, MasterCard, Carnet, and private label cards.
Needs
The company is looking to implement a solution to detect fraudulent transactions, reduce false positives, and minimize the risk impact by automating the process and providing dashboards and KPIs for decision-making. The goal is to create a single source of truth and analysis.
Solution
The proposed solution is the implementation of the Fraud Detector product developed and perfected by Nubiral.
The tool is a model created with Machine Learning, utilizing both supervised and unsupervised learning techniques.
Additionally, it would be integrated with the payment platform to stop suspicious transactions.
Results
After analysis, it was found that 0.06% of transactions were fraudulent, defining a threshold of 70% probability to classify them as such and take corrective actions in subsequent processes.
Casos de Éxito Data
15.000 correos por mes, clasificación 100% automática: IA en hotelería
Cómo migrar la infraestructura crítica de una empresa líder en entretenimiento masivo a Azure sin que los usuarios lo notaran — y sin un solo incidente.
Casos de Éxito Cloud
Modernización infraestructura crítica en Azure: Cero incidentes, 2M usuarios
Cómo migrar la infraestructura crítica de una empresa líder en entretenimiento masivo a Azure sin que los usuarios lo notaran — y sin un solo incidente.
Casos de Éxito Cloud Native Dev
Del crecimiento cloud descontrolado a un ahorro de hasta 50% con FinOps
Cómo una entidad financiera líder en Argentina transformó un crecimiento de gasto AWS del 896% en 13 meses en un plan FinOps con ahorro potencial de hasta el 50%.
