Meta-Learning Based Recommendation of Ensemble Methods for Gene Expression Classification

Title: Meta-Learning Based Recommendation of Ensemble Methods for Gene Expression Classification

Authors: José Gilberto Vasconcelos Júnior, Bruno Feres de Souza, André Carlos Ponce de Leon Ferreira de Carvalho

Abstract: For the past decade, microarray technology has been used to provide medical scientists a deeper understanding of diverse molecular phenomena. One of its most prominent applications is the identification of class membership of tissue samples based on their genetic profiles. For this task, Machine Learning algorithms have been commonly employed. In this paper, we present a meta-learning approach that recommends a suitable ensemble method for gene expression classification. Due to the nature of data considered, providing accurate recommendation is not trivial. Despite of that, our approach managed to outperform a baseline method, making room for new research directions.

Key-words: Gene expression data, Meta-learning, Ensemble

Pages: 6

DOI code: 10.21528/CBIC2019-107

PDF file: CBIC2019-107.pdf

BibTeX file: CBIC2019-107.bib