Augmented Functional Random Forests: Classifier Construction and Unbiased Functional Principal Components Importance through Ad-Hoc Conditional Permutations

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Maturo, Fabrizio, Porreca, Annamaria
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910575163867136
author Maturo, Fabrizio
Porreca, Annamaria
author_facet Maturo, Fabrizio
Porreca, Annamaria
contents This paper introduces a novel supervised classification strategy that integrates functional data analysis (FDA) with tree-based methods, addressing the challenges of high-dimensional data and enhancing the classification performance of existing functional classifiers. Specifically, we propose augmented versions of functional classification trees and functional random forests, incorporating a new tool for assessing the importance of functional principal components. This tool provides an ad-hoc method for determining unbiased permutation feature importance in functional data, particularly when dealing with correlated features derived from successive derivatives. Our study demonstrates that these additional features can significantly enhance the predictive power of functional classifiers. Experimental evaluations on both real-world and simulated datasets showcase the effectiveness of the proposed methodology, yielding promising results compared to existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2408_13179
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Augmented Functional Random Forests: Classifier Construction and Unbiased Functional Principal Components Importance through Ad-Hoc Conditional Permutations
Maturo, Fabrizio
Porreca, Annamaria
Machine Learning
Statistics Theory
Methodology
62G05, 62J99, 62H30, 68T05, 68T20
G.3; I.2.6; I.5.2; I.5.1
This paper introduces a novel supervised classification strategy that integrates functional data analysis (FDA) with tree-based methods, addressing the challenges of high-dimensional data and enhancing the classification performance of existing functional classifiers. Specifically, we propose augmented versions of functional classification trees and functional random forests, incorporating a new tool for assessing the importance of functional principal components. This tool provides an ad-hoc method for determining unbiased permutation feature importance in functional data, particularly when dealing with correlated features derived from successive derivatives. Our study demonstrates that these additional features can significantly enhance the predictive power of functional classifiers. Experimental evaluations on both real-world and simulated datasets showcase the effectiveness of the proposed methodology, yielding promising results compared to existing methods.
title Augmented Functional Random Forests: Classifier Construction and Unbiased Functional Principal Components Importance through Ad-Hoc Conditional Permutations
topic Machine Learning
Statistics Theory
Methodology
62G05, 62J99, 62H30, 68T05, 68T20
G.3; I.2.6; I.5.2; I.5.1
url https://arxiv.org/abs/2408.13179