Método Híbrido Multi-Filtro-Multi-Wrapper Utilizado para la Selección de Genes de Bases de Datos Biomédicas. (Hybrid Multiple-Filter-Multiple-Wrapper Method Used for Gene Selection from Biomedical Database.)

Luis Alberto Hernández (luisahm@itamail.itapizaco.edu.mx)2, Edmundo Bonilla (edbonn@itapizaco.edu.mx)1, Roberto Morales (moralescaporal@itapizaco.edu.mx)2, José Crispín Hernández (josechh@yahoo.com)2, Ángel Diaz (isc.diazpacheco@gmail.com)2


1Instituto Tecnologico de Apizaco
2Instituto Tecnológico de Apizaco

This paper appears in: Revista IEEE América Latina

Publication Date: Feb. 2013
Volume: 11,   Issue: 1 
ISSN: 1548-0992


Abstract:
In this paper, it is proposed a hybrid multiple-filter/multiple-wrapper method for selection and classification of DNA-microarray data. The wrapper algorithm GA/TS/SVM, combines relevant genes obtained of five data filtering methods, to perform an efficiently guided search of a subset gene of the bases obtained from the DNA-microarray technology. In the experimental results, it is observed that the method GA/TS/SVM works accurately, and achieves a lower classification error rate using a smaller number of selected genes than other methods reported in literature.

Index Terms:
Genetic Algorithm, Tabu Search, Support Vector Machine, Data Filtering, Databases,   


Documents that cite this document
This function is not implemented yet.


[PDF Full-Text (793)]