Authors: Maria Del Rosario Suarez, Jose R. Villar, Javier Grande
Addresses: Department of Computer Science, University of Oviedo, Edif. Departamental 1, Campus de Viesques s/n, 33204 Gijon, Asturias, Spain. ' Department of Computer Science, University of Oviedo, Edif. Departamental 1, Campus de Viesques s/n, 33204 Gijon, Asturias, Spain. ' Software Engineering Department, Indra Sistemas, S.A., Moises de Leon 57 24006, Leon, Spain
Abstract: Attempting to obtain a classifier or a model from datasets could be a cumbersome task, specifically when using datasets of high dimensionality. The larger the amount of features the higher the complexity of the problem and the longer the time that is expended in generating the outcome (the classifier or the model). Feature selection has been proved as a good technique for choosing features that best describe the system under certain criteria or measure. There are several different approaches for feature selection, but to our knowledge, there are not many different approaches when feature selection is involved with imprecise data and genetic fuzzy systems. In this paper, a feature selection method based on the fuzzy mutual information is proposed. The outlined method is valid for classifying problems when expertise partitioning is given, and it represents the base of future work including the use of imprecise data.
Keywords: mutual information; classification; feature selection; imprecise data; genetic fuzzy systems; descriptive reasoning; genetic algorithms; fuzzy logic; expertise partitioning.
International Journal of Reasoning-based Intelligent Systems, 2010 Vol.2 No.2, pp.133 - 141
Available online: 30 Aug 2010Full-text access for editors Access for subscribers Purchase this article Comment on this article