Extended Abstract (620K)2.5
A storm-type classifier using support vector machines and fuzzy logic
Jenny Abernethy, NCAR/RAL, Boulder, CO; and J. K. Williams
This paper details our methodology for entry in the 2008 American Meteorological Society's 6th Conference on Artificial Intelligence Applications to Environmental Science AI competition. We trained probabilistic Support Vector Machine models with several separations of a multi-category storm type data set. We experimented with two fuzzy logic approaches to combine the model outputs into a coherent prediction for our contest entry.
Session 2, Forecasting contest submissions (Participants will present the results of their model forecasts of the posted dataset)
Tuesday, 22 January 2008, 1:30 PM-3:15 PM, 205
Previous paper Next paper
Recorded presentation