O. Ibarrondo Olagüenaga, S. Pérez Fernández
There are several scenarios in which functional data are collected. This is the case of normalized fluorescence versus temperature, an analytical technique commonly used to find some particular genetic features. Our goal is to try to detect lactose intolerance using that measure, this is therefore a supervised binary classification problem from a functional marker. There are different methods to deal with functional data classification, both parametric and nonparametric, many of the last ones already implemented in the R package fda.usc. We compare those methods with a different approach relying on Support Vector Machines methodology on the basis of the lactose intolerance dataset, providing a clear implementation of the algorithm also in R software.
Palabras clave / Keywords: functional data, classification, Support Vector Machines
Programado
Pósteres I
30 de mayo de 2018 15:30
Zona EXPO
