A local discretization of continuous data for lattices: Technical aspects - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2011

A local discretization of continuous data for lattices: Technical aspects

Résumé

Since few years, Galois lattices (GLs) are used in data mining and defi ning a GL from complex data (i.e. non binary) is a recent challenge. Indeed GL is classically defi ned from a binary table (called context), and therefore in the presence of continuous data a discretization step is generally needed to convert continuous data into discrete data. Discretization is classically performed before the GL construction in a global way. However, local discretization is reported to give better classifi cation rates than global discretization when used jointly with other symbolic classi cation methods such as decision trees (DTs). Using a result of lattice theory bringing together set of objects and specifi c nodes of the lattice, we identify subsets of data to perform a local discretization for GLs. Experiments are performed to assess the effi ciency and the effectiveness of the proposed algorithm compared to global discretization.
Fichier non déposé

Dates et versions

hal-00634570 , version 1 (21-10-2011)

Identifiants

  • HAL Id : hal-00634570 , version 1

Citer

Nathalie Girard, Karell Bertet, Muriel Visani. A local discretization of continuous data for lattices: Technical aspects. International Conference on Concept Lattices and Their Applications, Oct 2011, Nancy, France. pp.409 - 412. ⟨hal-00634570⟩

Collections

L3I UNIV-ROCHELLE
28 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More