Autoencoder-based Attribute Noise Handling Method for Medical Data - Archive ouverte HAL Access content directly
Proceedings Year : 2023

Autoencoder-based Attribute Noise Handling Method for Medical Data

Abstract

Medical datasets are particularly subject to attribute noise, that is, missing and erroneous values. Attribute noise is known to be largely detrimental to learning performances. To maximize future learning performances it is primordial to deal with attribute noise before any inference. We propose a simple autoencoder-based preprocessing method that can correct mixed-type tabular data corrupted by attribute noise. No other method currently exists to handle attribute noise in tabular data. We experimentally demonstrate that our method outperforms both state-of-the-art imputation methods and noise correction methods on several real-world medical datasets.
Fichier principal
Vignette du fichier
main.pdf (188.56 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03696250 , version 1 (15-06-2022)

Identifiers

Cite

Thomas Ranvier, Haytham Elgazel, Emmanuel Coquery, Khalid Benabdeslem. Autoencoder-based Attribute Noise Handling Method for Medical Data. 1793, Springer Nature Singapore, pp.212-223, 2023, Communications in Computer and Information Science, ⟨10.1007/978-981-99-1645-0_18⟩. ⟨hal-03696250⟩
61 View
99 Download

Altmetric

Share

Gmail Facebook X LinkedIn More