Geodesic
Optimal discretization and selection of features by association rates of joint distributions
Santoni, Daniele1 ; Weitschek, Emanuel1, 2 ; Felici, Giovanni1
1 Institute for System Analysis and Computer Science “Antonio Ruberti”, National Research Council of Italy, Via dei Taurini 19, 00185 Rome, Italy
2 Department of Engineering, Uninettuno International University, Corso Vittorio Emanuele II, 39, 00186 Rome, Italy.
RAIRO - Operations Research - Recherche Opérationnelle, Special issue: Research on Optimization and Graph Theory dedicated to COSI 2013 / Special issue: Recent Advances in Operations Research in Computational Biology, Bioinformatics and Medicine, Tome 50 (2016) no. 2, pp. 437-449

Voir la notice de l'article provenant de la source Numdam

In this paper we propose a new method to measure the contribution of discretized features for supervised learning and discuss its applications to biological data analysis. We restrict the description and the experiments to the most representative case of discretization in two intervals and of samples belonging to two classes. In order to test the validity of the method, we measured the abundance of different explanatory models that can be derived from a given set of binary features. We compare the performances of our algorithm with those of popular feature selection methods, over three different publicly available gene expression data sets. The results of the comparison are in favour of the proposed method.

Reçu le :
Accepté le :
DOI : 10.1051/ro/2015045
Classification : 62H30
Keywords: Features selection, discretization, data mining

Santoni, Daniele 1 ; Weitschek, Emanuel 1, 2 ; Felici, Giovanni 1

1 Institute for System Analysis and Computer Science “Antonio Ruberti”, National Research Council of Italy, Via dei Taurini 19, 00185 Rome, Italy
2 Department of Engineering, Uninettuno International University, Corso Vittorio Emanuele II, 39, 00186 Rome, Italy. 
@article{RO_2016__50_2_437_0,
     author = {Santoni, Daniele and Weitschek, Emanuel and Felici, Giovanni},
     title = {Optimal discretization and selection of features by association rates of joint distributions},
     journal = {RAIRO - Operations Research - Recherche Op\'erationnelle},
     pages = {437--449},
     publisher = {EDP-Sciences},
     volume = {50},
     number = {2},
     year = {2016},
     doi = {10.1051/ro/2015045},
     zbl = {1341.62188},
     mrnumber = {3479881},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.1051/ro/2015045/}
}
TY  - JOUR
AU  - Santoni, Daniele
AU  - Weitschek, Emanuel
AU  - Felici, Giovanni
TI  - Optimal discretization and selection of features by association rates of joint distributions
JO  - RAIRO - Operations Research - Recherche Opérationnelle
PY  - 2016
SP  - 437
EP  - 449
VL  - 50
IS  - 2
PB  - EDP-Sciences
UR  - http://geodesic.mathdoc.fr/articles/10.1051/ro/2015045/
DO  - 10.1051/ro/2015045
LA  - en
ID  - RO_2016__50_2_437_0
ER  - 
%0 Journal Article
%A Santoni, Daniele
%A Weitschek, Emanuel
%A Felici, Giovanni
%T Optimal discretization and selection of features by association rates of joint distributions
%J RAIRO - Operations Research - Recherche Opérationnelle
%D 2016
%P 437-449
%V 50
%N 2
%I EDP-Sciences
%U http://geodesic.mathdoc.fr/articles/10.1051/ro/2015045/
%R 10.1051/ro/2015045
%G en
%F RO_2016__50_2_437_0
Santoni, Daniele; Weitschek, Emanuel; Felici, Giovanni. Optimal discretization and selection of features by association rates of joint distributions. RAIRO - Operations Research - Recherche Opérationnelle, Special issue: Research on Optimization and Graph Theory dedicated to COSI 2013 / Special issue: Recent Advances in Operations Research in Computational Biology, Bioinformatics and Medicine, Tome 50 (2016) no. 2, pp. 437-449. doi: 10.1051/ro/2015045

Cité par Sources :