Feature Selection by Conformal Predictor - Archive ouverte HAL Access content directly
Conference Papers Year : 2011

Feature Selection by Conformal Predictor

(1) , (1) , (1) , (1)
1

Abstract

In this work we consider the problem of feature selection in the context of conformal prediction. Unlike many conventional machine learning methods, conformal prediction allows to supply individual predictions with valid measure of confidence. The main idea is to use confidence measures as an indicator of usefulness of different features: we check how many features are enough to reach desirable average level of confidence. The method has been applied to abdominal pain data set. The results are discussed.
Fichier principal
Vignette du fichier
978-3-642-23960-1_51_Chapter.pdf (83.36 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01571493 , version 1 (02-08-2017)

Licence

Attribution - CC BY 4.0

Identifiers

Cite

Meng Yang, Ilia Nouretdinov, Zhiyuan Luo, Alex Gammerman. Feature Selection by Conformal Predictor. 12th Engineering Applications of Neural Networks (EANN 2011) and 7th Artificial Intelligence Applications and Innovations (AIAI), Sep 2011, Corfu, Greece. pp.439-448, ⟨10.1007/978-3-642-23960-1_51⟩. ⟨hal-01571493⟩
196 View
64 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More