Machine Learning Research Takes Aim At Fraud

Machine Learning Research Takes Aim At Fraud

by February 16, 2017
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  

Swiss Fintech Company NetGuardians and the School of Engineering and Management Vaud (HEIG-VD) have joined forces in a Made-in-Switzerland research project aimed at taking machine learning and artificial intelligence (AI) technology in financial fraud detection to the next level. The project is being supported by the Swiss Commission for Technology and Innovation (CTI).

Bringing together leading-edge industry and academic strengths, the collaboration will further develop NetGuardians’ current real-time fraud detection technologies that use machine learning for superior analytics across all channels and banking systems.

NetGuardians will work with the Institute for Information and Communication Technologies (IICT), an interdisciplinary applied research institute for real-world IT challenges, based at the technology-focused university HEIG-VD.

“At HEIG-VD, cross-disciplinary expertise brings varied academic perspectives to industry challenges,” says HEIG-VD Prof. Stephan Robert of the IICT. “To know industry needs better, and to transfer technology to market faster, working with industry leaders likeNetGuardians is key to our strategy.
He adds: “Together we aim to harness the potential of machine learning and AI so financial institutions can drastically reduce current manual procedures in place and mitigate risks, detecting previously unimaginable fraud patterns.”

Complementing NetGuardians’ internal R&D strategies, HEIG-VD will bring new algorithmic approaches and academic knowledge on machine learning and AI to further optimize fraud detection, prevention and operational control.

In addition, a criminologist from the University of Lausanne will participate in the project, notes NetGuardians Head of R&D Jérôme Kehrli; “NetGuardians’ technology DNA is rooted in user behavior analytics.

For this, understanding the fraudster’s psychology improves precision of scoring to identify priority cases for fraud detection. Combined with advanced data analytics, criminology helps predict crime – and therefore prevent it.

“The current industry paradigm for machine learning and AI capabilities in fraud detection remains primordial as there are too many false positives and too many valid transaction blockings. Our objective with this research project is to develop better tools that help banks achieve their ultimate goals – deliver an exceptional customer experience, strengthen operational efficiency, prevent emerging risks, and reduce costs.”

Print Friendly
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •