Digital payments firm Payoneer has deployed a data science platform called Iguazio to prevent transaction fraud for its four million customers.
Iguazio will leverage real-time machine learning and predictive analytics. The move is said to change Payoneer’s fraud detection approach to a prevention method.
The payments company uses algorithms that monitor various parameters such as account creation time and name change to detect fraud in complex networks.
However, the detection approach can help block users only after the fraud has occurred. With the new Iguazio solution, Payoneer can use the same machine learning techniques for real-time data.
The new solution is expected to instantly prevent fraud and money laundering. This will be made possible via predictive machine learning models that continuously look for suspicious patterns.
Payoneer added that this collaboration was facilitated by Belocal.

US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataPayoneer vice-president of Corporate Security and Global IT Operations Yaron Weiss said: “We’ve tackled one of our most elusive challenges with real-time predictive models, making fraud attacks almost impossible on Payoneer.
“With Iguazio’s Data Science Platform, we built a scalable and reliable system which adapts to new threats and enables us to prevent fraud with minimum false positives.”
Iguazio CEO Asaf Somekh said: “We’re glad to see Payoneer accelerating its ability to develop new machine learning-based services, increasing the impact of data science on the business.”
Iguazio has a low latency serverless framework, real-time multi-model data engine and Python eco-system, said Payoneer.
Recently, Iguazio raised $24m from Samsung SDS and Kensington Capital Partners, among other investors. The company will use the funding for product expansion and market growth.
Last month, Payoneer agreed to buy optile, a provider of open payment orchestration platform.