https://web.archive.org/web/20150201000000*/https://www.sandia.gov/~wpk/
Philip Kegelmeyer (Ph.D, Stanford, Information Systems Lab, 1985) is a Senior Scientist at Sandia National Laboratories in Livermore, CA. His current interests include machine learning and ensemble methods (especially as applied to sparse, sequence data) and adversarial machine learning issues.
In prior work, he was the Principal Investigator for a large (forty part-time staff) "Grand Challenge" internal research and development project devoted to network analysis. The project concluded in October of 2010; see the Networks Grand Challenge website for the final report and all related publications.
Prior to the NGC he led the Advanced Simulation Computing Data Discovery Program, devoted to search in, and characterization of, petascale scientific simulation data.
He has twenty years experience inventing, tinkering with, and quantitatively improving supervised machine learning algorithms (particularly ensemble methods), including published and continuing investigations into how to accurately and statistically significantly compare such algorithms. His work has resulted in over fifty refereed publications, two patents, and commercial software licenses. His recent research activities continue his interest in ensemble methods, particularly in the context of supervised learning in text.