Dynamic Attribution Platform for Predicting and Preventing Criminal or Fraudulent Activity
Data security breaches and fraud are becoming more and more frequent, and increasingly sophisticated methods are being used to steal data or cause targeted damage, hindering efforts to prevent attacks.
Researchers at the University of Tennessee have developed a novel method for predicting undesirable events using similarity-based information retrieval software. This method uses multivariate statistical analysis and principal component analysis for indexing object attributes. The software can predict if an event is likely to occur and identify who/where the fraudulent activity is coming from, thereby enabling precautionary measures to be taken to avert the attack.
- Can be easily tailored for any industry
- Minimizes financial losses
- Improves decision making
- Mitigates risk
- Enhances privacy
- Increases customer confidence
- Predict, detect, and prevent criminal/fraudulent activity in
- Health care
- Law enforcement, etc.
- Regulatory compliance
- Risk management
- U.S. 8,375,032
- U.S. 8,392,418
- U.S. 8,396,870
- U.S. 8,713,019
- U.S. 8,762,379
- U.S. 8,775,427
- U.S. 8,775,428
Dr. David Icove is a UL Professor of Practice in the Department of Electrical Engineering and Computer Science at UT. His research focuses on forensic engineering and high-performance computer modeling of fires and explosions; and cyberterrorism, intrusion detection, and computer security.
Dr. Doug Birdwell is a Professor Emeritus in the Department of Electrical and Computer Engineering at UT. His research expertise includes control systems, information processing, high-performance databases, data mining, and bioinformatics.
Dr. Tsewei Wang is an Associate Professor Emeritus of Chemical and Biomolecular Engineering at UT. Her research interests include data mining, process monitoring and fault detection using multivariate statistical methods; and bioinformatics, especially in the field of DNA forensics, using DNA for human identification.