Hybrid methodology based on Bayesian optimization and GA-PARSIMONY to search for parsimony models by combining hyperparameter optimization and feature selection
Authors:
- F.J. Martinez-De-PisonF.J. Martinez-De-Pison
- R. Gonzalez-SendinoRGR. Gonzalez-Sendino
- A. AldamaAAA. Aldama
- J. Ferreiro-CabelloJFJ. Ferreiro-Cabello
- E. Fraile-GarciaEFE. Fraile-Garcia
- 1 August 2019
- journal article
- research article
- Published by Elsevier in Neurocomputing
- Vol. 354, 20-26
Abstract
No abstract availableKeywords
Funding Information
- Banco Santander (APPI16/05, APPI17/04, REGI2018/43)
- MINECO (UNLR-094E-2C-225)
This publication has 27 references indexed in Scilit:
- Modified binary PSO for feature selection using SVM applied to mortality prediction of septic patientsApplied Soft Computing, 2013
- Recent trends in intelligent data analysisNeurocomputing, 2013
- Evolutionary algorithm characterization in real parameter optimization problemsApplied Soft Computing, 2013
- Meta-learning for evolutionary parameter optimization of classifiersMachine Learning, 2012
- Spectral and Wavelet-based Feature Selection with Particle Swarm Optimization for Hyperspectral ClassificationJournal of Software, 2011
- A distributed PSO–SVM hybrid system with feature selection and parameter optimizationApplied Soft Computing, 2008
- ESVM: Evolutionary support vector machine for automatic feature selection and classification of microarray dataBioSystems, 2007
- Parsimonious additive modelsComputational Statistics & Data Analysis, 2007
- Greedy function approximation: A gradient boosting machine.The Annals of Statistics, 2001
- Improving the accuracy of decision tree induction by feature preselectionApplied Artificial Intelligence, 2001