The goal of this adapt and apply solution is to enable investigator teams to understand how Dataiku can be used to leverage established insights and rules for credit card fraud detection modeling within a robust and full-featured data science platform, while easily incorporating new machine learning approaches and ensuring real-time alerts management. More details on the specifics of the solution can be found on the knowledge base.
The total value of fraudulent transactions using cards issued within SEPA and acquired worldwide amounted to €1.80 billion in 2018¹, an increase of 13% year on year. This trend calls for continuous fraud monitoring and vigilance, both to limit banks P&L impact but also to enhance customer trust
Fraud detection rules are complex and well-established though often are based on business rules only. Enhancing set-ups with machine learning integration opens opportunities for increased efficiency in better detecting fraudulent behaviors and maximizing focus.
Providing a unified space for teams to manage business rules alongside machine learning approaches, and allowing for sandbox experimentation and enterprise-grade productionization, ensures the gains from machine learning are realized, without losing established success through existing approaches.
- Unifying business and ML rules ensures that existing approaches are leveraged to their full potential while adding incremental value via ML.
- Business-friendly explainable AI allows for rapid and intelligent review of machine learning models, ensuring all teams are confident and contribute expertise effectively.
- Integrated model re-evaluation and streamlined redeployment allow for models to be quickly retrained while retaining full control over production decisions.
- Integrated alerts monitoring and powerful case management connectivity allows for effective alert assessment and assignment
Business & ML Rules
Develop a comprehensive fraud detection approach incorporating business rules & ML rules via a unified scoring model.
Powerful data exploration
Explore and better understand the impact of existing and newly developed rules and models on alert generation, and the underlying behavior of transactions data.
Business-interpretable model insights
Alert detection and scoring models include easily understood insights into potentially relevant variables to leverage when creating new or revising existing business rules.
Streamlined model approval and review
Easily review drift of alert generation and scoring models and data to streamline approval, and use the same simple process to re-evaluate the model as time passes.
Easy deployment of refreshed models
Use integrated and simple model retraining to quickly and transparently deploy updated models, while maintaining full governance and control at all times.
End-to-end process enhancement
Leverage the full capabilities of Dataiku to enhance your existing fraud detection and alert generations process end-to-end: combine agile business rules with machine learning to optimize rules generation with full explainability and governance.