The 2008 Artificial Intelligence Competition
暂无分享,去创建一个
The Artificial Intelligence Committee of the AMS conducted a classification competition to correctly classify storms into one of four types based on radar-derived characteristics. This competition, which was sponsored by Weather Decision Technologies, received seven entries, three of which beat the baseline decision tree used in the original work that used the dataset. We describe the dataset for automated storm type classification. The dataset is the result of a clustering program run on multi-radar reflectivity images and properties of those clusters computed by a set of severe weather algorithms. The labels on the dataset, created by manual identification, identify each cluster as being one of four categories: supercells, convective lines, pulse storms or non-organized cells. The ranking was carried out using the True Skill Statistic.
[1] Donald W. Burgess,et al. Tornado and Severe Thunderstorm Warning Forecast Skill and its Relationship to Storm Type , 2007 .
[2] Robert M. Rabin,et al. Multiscale storm identification and forecast , 2003 .
[3] J. Ross Quinlan,et al. C4.5: Programs for Machine Learning , 1992 .
[4] Travis M. Smith,et al. The Warning Decision Support System–Integrated Information , 2007 .