Machine learning method cuts fraud detection costs by generating accurate labels from imbalanced datasets
- Florida Atlantic University researchers developed a novel machine learning method for fraud detection.
- Existing fraud detection methods struggle with severely imbalanced datasets containing few fraud instances.
- The team tested the method on credit card data and Medicare Part D claims from 2013-2019.
- Mary Anne Walauskis said their method refines labels to minimize fraud labels and false positives.
- The new method promises more accessible fraud detection and safeguards financial and health systems.
7 Articles
7 Articles
Machine learning method cuts fraud detection costs by generating accurate labels from imbalanced datasets
Fraud is widespread in the United States and increasingly driven by technology. For example, 93% of credit card fraud now involves remote account access, not physical theft. In 2023, fraud losses surpassed $10 billion for the first time.
How next-generation AI and data clusters are pioneering fraud defense - WorldNL Magazine
Fraud prevention is undergoing massive shifts as organizations strive to stay ahead of increasingly sophisticated bad actors. Thanks to advancements, such as artificial intelligence (AI) and machine learning (ML), that democratize access to enable fraud at speed and scale, traditional, rules-based systems – long the industry standard – are proving insufficient in the wake of such emerging technologies. As bad actors find new and innovative ways …
Fraud Detection Transformed: Researchers Harness Machine Learning for
In a groundbreaking advancement in the realm of fraud detection, researchers from Florida Atlantic University’s College of Engineering and Computer Science have harnessed the power of machine learning to tackle the ever-evolving challenges of fraud in health care and finance. As fraud continues to escalate—costing the U.S. economy billions every year—this innovative method represents a significant step towards more effective and efficient identi…
Coverage Details
Bias Distribution
- 100% of the sources are Center
To view factuality data please Upgrade to Premium
Ownership
To view ownership data please Upgrade to Vantage