Data-Driven Anomaly Detection | Nikunj Oza | Talks at Google

This talk will describe recent work by the NASA Data Sciences Group on
data-driven anomaly detection applied to air traffic control over Los Angeles, Denver, and New York. This data mining approach is designed to discover operationally significant flight anomalies, which were not pre-defined. These methods are complementary to traditional exceedance-based methods, in that they are more likely to yield false alarms, but they are also more likely to find previously-unknown anomalies. We discuss the discoveries that our algorithms have made that exceedance-based methods did not identify.

Nikunj Oza is the leader of the Data Sciences Group at NASA Ames Research Center. He also leads a NASA project team which applies data mining to aviation safety. Dr. Ozaąs 40+ research papers represent his research interests which include data mining, machine learning, anomaly detection, and their applications to Aeronautics and Earth Science. He received the Arch T. Colwell Award for co-authoring one of the five most innovative technical papers selected from 3300+ SAE technical papers in 2005. His data mining team received the 2010 NASA Aeronautics Research Mission Directorate Associate Administratorąs Award for best technology achievements by a team. He received his B.S. in Mathematics with Computer Science from MIT in 1994, and M.S. (in 1998) and Ph.D. (in 2001) in Computer Science from the University of California at Berkeley.