Advances in Feature Selection with Mutual Information
The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps fighting the curse of dimensionality, improving the performances of prediction or classification methods, and interpreting the application. In a nonlinear context, the mutual information is widely used as relevance criterion for features and sets of features. Nevertheless, it suffers from at least three major limitations: mutual information estimators depend on smoothing parameters, there is no theoretically justified stopping criterion in the feature selection greedy procedure, and the estimation itself suffers from the curse of dimensionality. This chapter shows how to deal with these problems. The two first ones are addressed by using resampling techniques that provide a statistical basis to select the estimator parameters and to stop the search procedure. The third one is addressed by modifying the mutual information criterion into a measure of how features are complementary (and not only informative) for the problem at hand.
Unable to display preview. Download preview PDF.
- 3.Dijck, G.V., Hulle, M.M.V.: Speeding up the wrapper feature subset selection in regression by mutual information relevance and redundancy analysis. In: ICANN 2006: International Conference in Aritificial Neural Networks, Athens, Greece, September 2006, pp. 31–40 (2006) (submitted) Google Scholar
- 4.François, D., Krier, C., Rossi, F., Verleysen, M.: Estimation de redondance pour le clustering de variables spectrales. In: Agrostat 2008, 10èmes journées Européennes Agro-industrie et Méthodes statistiques, pp. 55–61. Louvain-la-Neuve, Belgium (2008)Google Scholar
- 7.Kozachenko, L.F., Leonenko, N.N.: Sample estimate of entropy of a random vector. Probl. Inf. Transm. 23, 95–101 (1987)Google Scholar
- 9.Krier, C., François, D., Rossi, F., Verleysen, M.: Feature clustering and mutual information for the selection of variables in spectral data. In: ESANN 2007, European Symposium on Artificial Neural Networks Advances in Computational Intelligence and Learning, pp. 157–162. Bruges, Belgium (2007)Google Scholar