Zaffran, Mme Margaux (2020) Clustering of rainfall extremes by coupling Kullback-Leibler divergence and machine learning algorithms PFE - Project Graduation, ENSTA.



Climate extremes have a strong societal impact, especially extreme rainfalls. Recent dramatic events, like Alex Storm that stroke northern Italy and South-East of France, emphasize the need of a good knowledge of extreme events to adapt infrastructures. Inherently suffering of too few data, this field would benefit a tool which creates homogeneous regions to improve the extreme studies. We propose a clustering method tailored for extreme and based on the probability distributions. Furthermore, we analyze its results on a daily rainfall data-set from Meteo-France.

Item Type:Thesis (PFE - Project Graduation)
Uncontrolled Keywords:clustering, Kullback-Leibler divergence, extreme values, Extreme Value Theory, rainfall, climate
Subjects:Mathematics and Applications
ID Code:8291
Deposited By:Margaux Zaffran
Deposited On:13 nov. 2020 09:00
Dernière modification:13 nov. 2020 09:00

Repository Staff Only: item control page