A Hybrid Knowledge Discovery Approach for Mining Predictive Biomarkers in Metabolomic Data - Archive ouverte HAL Access content directly
Conference Papers Year : 2016

A Hybrid Knowledge Discovery Approach for Mining Predictive Biomarkers in Metabolomic Data

Blandine Comte
Estelle Pujos-Guillot
Amedeo Napoli

Abstract

The analysis of complex and massive biological data issued from metabolomic analytical platforms is a challenge of high importance. The analyzed datasets are constituted of a limited set of individuals and a large set of features where predictive biomarkers of clinical outcomes should be mined. Accordingly, in this paper, we propose a new hybrid knowledge discovery approach for discovering meaningful predic-tive biological patterns. This hybrid approach combines numerical classifiers such as SVM, Random Forests (RF) and ANOVA, with a symbolic method, namely Formal Concept Analysis (FCA). The related experiments show how we can discover among the best potential predictive biomarkers of metabolic diseases thanks to specific combinations of clas-sifiers mainly involving RF and ANOVA. The visualization of predictive biomarkers is based on heatmaps while FCA is mainly used for visual-ization and interpretation purposes, complementing the computational power of numerical methods.
Fichier principal
Vignette du fichier
dg-etal-final-ecmlpkdd16.pdf (312.62 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01421011 , version 1 (21-12-2016)

Identifiers

Cite

Dhouha Grissa, Blandine Comte, Estelle Pujos-Guillot, Amedeo Napoli. A Hybrid Knowledge Discovery Approach for Mining Predictive Biomarkers in Metabolomic Data. ECML PKDD, Sep 2016, Riva del garda, Italy. pp.572 - 587, ⟨10.1007/978-3-319-46128-1_36⟩. ⟨hal-01421011⟩
448 View
314 Download

Altmetric

Share

Gmail Facebook X LinkedIn More