Analysis of Extreme Precipitation Events Using a Novel Data Mining Approach (original) (raw)
An innovative data mining approach is presented and applied to investigate the climatic causes of extreme climatic events. Our approach comprises two main steps of knowledge extraction, applied successively in order to reduce the complexity of the original data set. The goal is to identify a much smaller subset of climatic variables that might still be able to describe or even predict the extreme events. The first step applies a class comparison technique. The second step consists of a decision tree learning algorithm used as a predictive model to map the set of statistically most significant climate variables identified in the previous step to classes of precipitation intensity. The methodology is employed to the study the climatic causes of two extreme events occurred in Brazil the last decade: the Santa Catarina 2008 extreme rainfall tragedy and the Amazon droughts of 2005 and 2010. In both cases, our results are in good agreement with analyses published in the literature.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.