3.1 The 4th annual AMS Artificial Intelligence Forecasting Contest

Tuesday, 25 January 2011: 3:30 PM
2A (Washington State Convention Center)
Jennifer Abernethy, NCAR/RAL, Boulder, CO; and S. E. Haupt, A. Pelliccioni, and J. K. Williams

The AMS Committee on Artificial Intelligence Applications to Environmental Science organizes an annual contest to encourage students and professionals alike to learn more about computational intelligence techniques by applying them to real-world atmospheric science problems. This year's contest focuses on forecasting urban ozone levels in Rome. Based on a dataset of meteorological and pollutant measurements, contestants were asked to use any computational intelligence technique, or combination of techniques, to make predictions of ozone levels on specified days throughout a year (pollutant measurements were withheld from the dataset for these days). Prizes will be awarded to the contestant or contestants who make the most accurate (by value) ozone predictions and who most successfully predict severe ozone levels (by category). Complete details of the contest, including data, skill scores, and deadlines can be found at http://ai.metr.ou.edu/. This paper will introduce a conference session in which selected contestants will present their methods and results, followed by a short awards ceremony.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner