Bi-objective Genetic Algorithm for Feature Selection in Ensemble Systems
This paper presents the use of a bi-objective genetic algorithm to select attributes for an ensemble system. This is achieved by using this technique to simultaneously maximize the individual diversity of the base classifiers and the group diversity of an ensemble system. In order to evaluate the possible solutions obtained by this technique, two filter-based evaluation criteria will be used. Filter-based criteria were chosen because they are independent of the learning algorithm and have a low computational cost.
KeywordsEnsemble systems Genetic Algorithms Feature Selection
Unable to display preview. Download preview PDF.
- 1.Yu, S.: Feature selection and classifier ensembles: A study on hyperspectral remote sensing data. PhD thesis. University of Antwerp, Antwerp (2003)Google Scholar
- 3.Opitz, D.W.: Feature selection for ensembles. In: Conference on Artificial Intelligence, AAAI. American Association for Artificial Intelligence, pp. 379–384 (1999)Google Scholar
- 4.Canuto, A.M.P., Abreu, M.C.C., de Melo Oliveira, L., Xavier Jr., J.C., de M. Santos, A.: Investigating the influence of the choice of the ensemble members in accuracy and diversity of selection-based and fusion-based methods for ensembles. Pattern Recognition Letters 28(4), 472–486 (2007)CrossRefGoogle Scholar
- 5.Kuncheva, L.I.: Combining Pattern Classifiers: Methods and Algorithms. Wiley Interscience (2004)Google Scholar
- 7.Frank, A., Asuncion, A.: UCI machine learning repository (2010), http://archive.ics.uci.edu/ml
- 9.Durillo, J., Nebro, A., Luna, F., Dorronsoro, B., Alba, E.: jmetal: a java framework for developing multi-objective optimization metaheuristics. Department of Languages and Computer Sciences, Tech. Rep. (2006)Google Scholar
- 10.Zitzler, E., Laumanns, M., Thiele, L.: Spea2: Improving the strength pareto evolutionary algorithm. Department of Electrical Engineering, Swiss Federal Institute of Technology, Tech. Rep. (2001)Google Scholar