Blog: Innovative Technique Can Spot Errors in Key Technological Systems

Innovative Technique Can Spot Errors in Key Technological Systems
National Science Foundation (01/27/10) Dybas, Cheryl

Researchers at the University of Colorado at Boulder (CU) and the U.S. National Center for Atmospheric Research (NCAR) have developed the Intelligent Outlier Detection Algorithm (IODA), a computational technique that can detect errors in sensitive technological systems. IODA is designed to perform quality control on time series data. The algorithm uses statistics, graph theory, image processing, and decision trees to compare incoming data to common patterns of failure. IODA identifies problems with data points and determines how a particular datum is inaccurate by treating the data as an image. "We thought that, by using image processing, we could teach the system to detect inconsistencies, somewhat like a person would," says NCAR's Andrew Weekley. CU's Kent Goodrich says their approach "is a radical departure from the usual techniques found in the time series literature."

View Full Article