Consistency of Derivative Based Functional Classifiers on Sampled Data

Fabrice Rossi (1) , Nathalie Villa-Vialaneix (2)

1 AxIS - Usage-centered design, analysis and improvement of information systems
2 IMT - Institut de Mathématiques de Toulouse UMR5219

In some applications, especially spectrometric ones, curve classifiers achieve better performances if they work on the $m$-order derivatives of their inputs. This paper proposes a smoothing spline based approach that give a strong theoretical background to this common practice.

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Fabrice Rossi, Nathalie Villa-Vialaneix. Consistency of Derivative Based Functional Classifiers on Sampled Data. ESANN 2008, 16th European Symposium on Artificial Neural Networks, Apr 2008, Bruges, Belgium. pp.445-450. ⟨hal-00666609⟩