Kernels for time series of exponential decay/growth processes
Résumé
Many processes exhibit exponential behavior. When kernel-based machines are applied on this type of data, conventional kernels such as the Gaussian kernel are not appropriate. In this paper, we derive kernels adapted to time series of exponential decay or growth processes. We provide a theoretical study of these kernels, including the issue of universality. Experimental results are given on a case study: chlorine decay in water distribution systems.
Mots clés
one-class
cybersecurity
chemistry computing
chemical reactions
chlorine
exponential distribution
time series
exponential decay processes
growth processes
exponential behavior
kernel-based machines
conventional kernels
Time series analysis
Machine learning
Kernel
water distribution systems
chlorine decay
Gaussian kernel
Training
Kernel function
Chemicals
Temperature measurement
one-class classification
normalization
kernel methods
Index Terms-Kernel function
support vector machines
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|
Loading...