PIIH: Learning Prediction Intensity Interval model for Hurricanes
Hurricane intensity (maximum sustained wind speed) is one of the most important indicators for a hurricane's destructive power. Unfortunately, hurricane intensity is notoriously difficult to predict. At the IDA@SMU research lab, we apply a novel data stream modeling technique called TRACDS to improve the prediction of hurricane intensity. This new approach named PIIH dynamically models hurricane life cycle behavior using historic data and then applies these models to predict hurricane intensities up to 5 days in advance. What is completely new with this approach is the fact that it models the hurricane life cycle and, in addition to single value predictions, it also provides ranges (high to low) for the expected intensity.
- This research is featured in the article “Discovery: New Forecasting Algorithm Helps Predict Hurricane Intensity and Wind Speed” (Dec. 5, 2011) by the National Science Foundation.
- Vladimir Jovanovic, Margaret H. Dunham, Michael Hahsler, and Yu Su. Evaluating hurricane intensity prediction techniques in real time. In Third IEEE ICDM Workshop on Knowledge Discovery from Climate Data, Proceedings of the of the 2011 IEEE International Conference on Data Mining Workshops (ICDMW 2011). IEEE, 2011.
- Michael Hahsler and Margaret H. Dunham. Temporal structure learning for clustering massive data streams in real-time. In SIAM Conference on Data Mining (SDM11). SIAM, 2011.
- Yu Su, Sudheer Chelluboina, Michael Hahsler, and Margaret H Dunham, A New Data Mining Model for Hurricane Intensity Prediction, 2nd IEEE ICDM Workshop on Knowledge Discovery from Climate Data, Proceedings of the 2010 IEEE International Conference on Data Mining Workshops (ICDMW 2010). IEEE, 2010
We would like to thank James Franklin (Hurricane Specialist Unit, NHC, NOAA) and Dr. Mark Demaria (NESDIS Regional and Mesoscale Meteorology Branch, CIRA) for their help.
This work is based on TRACDS sponsored by the National Science Foundation under Grant No. IIS-0948893.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.