Supervised variational relevance learning, an analytic geometric feature selection with applications to omic datasets
Contribuinte(s) |
Universidade Estadual Paulista (UNESP) |
---|---|
Data(s) |
22/10/2015
22/10/2015
01/05/2015
|
Resumo |
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) We introduce Supervised Variational Relevance Learning (Suvrel), a variational method to determine metric tensors to define distance based similarity in pattern classification, inspired in relevance learning. The variational method is applied to a cost function that penalizes large intraclass distances and favors small interclass distances. We find analytically the metric tensor that minimizes the cost function. Preprocessing the patterns by doing linear transformations using the metric tensor yields a dataset which can be more efficiently classified. We test our methods using publicly available datasets, for some standard classifiers. Among these datasets, two were tested by the MAQC-II project and, even without the use of further preprocessing, our results improve on their performance. |
Formato |
705-711 |
Identificador |
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6977958 Ieee-acm Transactions On Computational Biology And Bioinformatics. Los Alamitos: Ieee Computer Soc, v. 12, n. 3, p. 705-711, 2015. 1545-5963 http://hdl.handle.net/11449/129779 http://dx.doi.org/10.1109/TCBB.2014.2377750 WOS:000356608100022 |
Idioma(s) |
eng |
Publicador |
Ieee Computer Soc |
Relação |
Ieee-acm Transactions On Computational Biology And Bioinformatics |
Direitos |
closedAccess |
Palavras-Chave | #Suvrel #Relevance Learning #Analytic metric learning #Proteomics #Metabolomics #Genomics #Feature selection #Distance learning |
Tipo |
info:eu-repo/semantics/article |