Constraint Selection-Based Semi-supervised Feature Selection.

Mohammed Hindawi (1) , Kais Allab (2) , Khalid Benabdeslem (2)

1 LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
2 GAMA - Graphes, Apprentissage et Multi-Agents

In this paper, we present a novel feature selection approach based on an efficient selection of pair wise constraints. This aims at selecting the most coherent constraints extracted from labeled part of data. The relevance of features is then evaluated according to their efficient locality preserving and chosen constraint preserving ability. Finally, experimental results are provided for validating our proposal with respect to other known feature selection methods.

Mots clés: feature selection; Dimensionality reduction; constraint selection; feature selection.

Domaines: Apprentissage [cs.LG]

Mohammed Hindawi, Kais Allab, Khalid Benabdeslem. Constraint Selection-Based Semi-supervised Feature Selection.. Cook, Diane J. and Pei, Jian and 0010, Wei Wang and Zaïane, Osmar R. and Wu, Xindong. ICDM, Dec 2011, Vancouver,BC, Canada. IEEE, pp.1080-1085, 2011, ⟨10.1109/ICDM.2011.42⟩. ⟨hal-00874960⟩