Research institute for
signals, systems and
computational intelligence

Feature Selection

Most machine learning problems deal with datasets in which every of the given examples consist of a large list of features. Usually, the set of features comes without knowledge about the discriminative information provided by each element. In this scenario, feature selection is essential to perform the classification task with reduced complexity and acceptable performance. A key issue is to define a criterion in order to rank the features, discarding those features that are less relevant, redundant, or noisy. This depends on the particular task, the classifier and the properties of the data. Another key issue comes from the infeasibility of evaluating all possible combinations, which is why an intelligent search strategy is required for finding an optimal subset. The objective of feature selection is the improvement of a machine learning model, either in terms of learning speed, computational complexity, simplicity, interpretability of the representation, and/or generalization capability.

Team members

Leandro Vignolo
Matias Gerard

Publications

Exploring feature extraction methods for infant mood classification   sinc_VAM19.doi sinc_VAM19.pdf
L. D. Vignolo, E. M. Albornoz, C. E. Martínez
AI Communications, Volume 3, Number 32, page 191–206 – 2019
Feature Set Optimisation for Infant Cry Classification   sinc_VAM18.pdf
L. D. Vignolo, E. M. Albornoz, C. E. Martínez
Advances in Artificial Intelligence – IBERAMIA 2018, page 455–466 – 2018
Automatic classification of Furnariidae species from the Paranaense Littoral region using speech-related features and machine learning   sinc_AVSL17.doi sinc_AVSL17.pdf
E. M. Albornoz, L. D. Vignolo, J. A. Sarquis, E. Leon
Ecological Informatics , Volume 38, page 39 – 49 – 2017
Evolutionary local improvement on genetic algorithms for feature selection   sinc_VG17.pdf
L. D. Vignolo, M. Gerard
Proceedings de XLIII CLEI – 46 JAIIO – 2017
Feature optimisation for stress recognition in speech   sinc_VPDRM16.doi sinc_VPDRM16.pdf
L. D. Vignolo, S. R. M. Prasanna, S. Dandapat, H. L. Rufiner, D. H. Milone
Pattern Recognition Letters, Volume 84, page 1–7 – dec 2016
Furnariidae species recognition using speech-related features and machine learning   sinc_VSLA16.pdf
L. D. Vignolo, J. A. Sarquis, E. Leon, E. M. Albornoz
17th Argentine Symposium on Artificial Intelligence (ASAI) – 45th JAIIO – sep 2016
Multi-objective optimisation of wavelet features for phoneme recognition   sinc_VRM16.doi VRM16.doi sinc_VRM16.pdf
L. D. Vignolo, H. L. Rufiner, D. H. Milone
IET Signal Processing, Volume 10, Number 6, page 685–691 – 2016
Multi-objective optimisation of wavelet features for phoneme recognition   sinc_VRM16a.pdf
L. D. Vignolo, H. L. Rufiner, D. H. Milone
17th Argentine Symposium on Artificial Intelligence (ASAI) – 45th JAIIO – nov 2016
Feature selection for face recognition based on multi-objective evolutionary wrappers   VMS13.doi sinc_VMS13.pdf
L. D. Vignolo, D. H. Milone, J. Scharcanski
Expert Systems with Applications, Volume 40, Number 13, page 5077–5084 – 2013
An Evolutionary Wrapper for Feature Selection in Face Recognition Applications   sinc_VMSB12.pdf
L. D. Vignolo, D. H. Milone, J. Scharcanski, C. Behaine
2012 IEEE International Conference on Systems, Man, and Cybernetics (IEEE SMC 2012) – 2012
All publications