Multi-Layer Perceptrons and Symbolic Data

Fabrice Rossi (1, 2) , Brieuc Conan-Guez (1, 3)

1 AxIS - Usage-centered design, analysis and improvement of information systems
2 CEREMADE - CEntre de REcherches en MAthématiques de la DEcision
3 LITA - Laboratoire d'Informatique Théorique et Appliquée

In some real world situations, linear models are not sufficient to represent accurately complex relations between input variables and output variables of a studied system. Multilayer Perceptrons are one of the most successful non-linear regression tool but they are unfortunately restricted to inputs and outputs that belong to a normed vector space. In this chapter, we propose a general recoding method that allows to use symbolic data both as inputs and outputs to Multilayer Perceptrons. The recoding is quite simple to implement and yet provides a flexible framework that allows to deal with almost all practical cases. The proposed method is illustrated on a real world data set.

Mots clés: Multi-Layer Perceptron; Symbolic Data; Interval Data

Domaines: Réseau de neurones [cs.NE]

Fabrice Rossi, Brieuc Conan-Guez. Multi-Layer Perceptrons and Symbolic Data. Diday, Edwin and Noirhomme-Fraiture, Monique. Symbolic Data Analysis and the SODAS Software, Wiley, pp.373-391, 2008. ⟨inria-00232878⟩