MD-ELM: Originally Mislabeled Samples Detection using OP-ELM Model

Anton Akusok, David Veganzones, Yoan Miche, Kaj-Mikael Björk, Philippe du Jardin, Eric Severin, Amaury Lendasse

Forskningsoutput: TidskriftsbidragArtikelVetenskapligPeer review

6 Citeringar (Scopus)

Sammanfattning

This paper proposes a methodology for identifying data samples that are likely to be mislabeled in a c-class classification problem (dataset). The methodology relies on an assumption that the generalization error of a model learned from the data decreases if a label of some mislabeled sample is changed to its correct class. A general classification model used in the paper is OP-ELM; it also provides a fast way to estimate the generalization error by PRESS Leave-One-Out. It is tested on two toy datasets, as well as on real life datasets for one of which expert knowledge about the identified potential mislabels has been sought.
OriginalspråkEngelska
Referentgranskad vetenskaplig tidskriftNeurocomputing
Volym159
UtgåvaJuly
Sidor (från-till)242-250
Antal sidor9
ISSN0925-2312
DOI
StatusPublicerad - 14.02.2015
MoE-publikationstypA1 Originalartikel i en vetenskaplig tidskrift

Nyckelord

  • 512 Företagsekonomi

Fingeravtryck

Fördjupa i forskningsämnen för ”MD-ELM: Originally Mislabeled Samples Detection using OP-ELM Model”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här