SEON is delighted to announce a new CatBoost blackbox machine learning update for its fraud-fighting solution.
By making highly accurate modelling accessible to everyone, CatBoost offers one of the best open-source machine learning interpretation tools available today. Now, the world’s fastest growing fraud prevention company, SEON is excited to announce the system has been incorporated into its industry-leading solution. Thanks to the update, more businesses can have access to enhanced efficiency and precision when fighting fraud.
Moving forward, SEON’s customers are now able to leverage the speed of its blackbox machine learning, alongside the transparency of whitebox machine learning scoring and rules. Once activated, the blackbox machine learning scores are layered into the scoring decision, powered by SEON’s scoring engine. The blackbox scores of each transaction are then used to create statistics, reports and modelling charts, which assess the risk of fraud.
With SEON’s latest update, users can calculate the probability of a transaction being fraudulent immediately. What’s more, the innovative system ‘learns’ at speed, meaning it doesn’t require any maintenance from the users’ end. Customer rules can also be applied to the system to drill down further and provide more transparency in how specific scores are determined. Furthermore, to refine accuracy even more quickly, users can also utilize SEON’s Label API within their systems to provide feedback that finesses the algorithm and decisions.
Speaking on the update, Jimmy Fong, Chief Commercial Officer at SEON commented: “We’ve updated our system with the best open-source machine learning interpretation tool available today. The new update uses a revolutionary blackbox machine learning system, which makes highly accurate modelling accessible to everyone. What’s more, in refining our system, we’re now able to provide another layer of machine-learning to support faster and accurate decision-making for our customers.
“The speed from the blackbox machine learning update complements our transparent whitebox machine learning scoring and rules. By combining them, we’re able to offer the best of both worlds: blazing speed and accuracy in detecting patterns, plus visibility into how machine learning decisions are made. Better still, the new update doesn’t interrupt existing SEON integrations, so users won’t need to make any development changes.”
Alongside the new blackbox machine learning update, SEON has also worked to improve the dashboard design of its system. Thanks to the update, the system now intuitively shows the important stats and data that customers most need. Additionally, SEON has updated the code under the hood of its service, so that its solution runs and displays faster. In turn, this also supports the company’s plans for an even bigger streamlined update soon.
SEON strives to help online businesses reduce the costs, time, and challenges faced due to fraud. With a real-time, flexible API, we collect all the relevant risk-related data points, and after connecting them, we provide a risk score that leverages data enrichment and machine learning.