Skip to Main content Skip to Navigation
Conference papers

Optimized Representation for Classifying Qualitative Data

Martine Cadot 1, * Alain Lelu 2, 3
* Corresponding author
1 ABC - Machine Learning and Computational Biology
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
2 KIWI - Knowledge Information and Web Intelligence
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : Extracting knowledge out of qualitative data is an ever-growing issue in our networking world. Opposite to the widespread trend consisting of extending general classification methods to zero/one-valued qualitative variables, we explore here another path: we first build a specific representation for these data, respectful of the non-occurrence as well as presence of an item, and making the interactions between variables explicit. Combinatorics considerations in our Midova expansion method limit the proliferation of itemsets when building level k+1 on level k, and limit the maximal level K. We validate our approach on three of the public access datasets of University of California, Irvine, repository: our generalization accuracy is equal or better than the best reported one, to our knowledge, on Breast Cancer and TicTacToe datasets, honorable on Monks-2 near-parity problem.
Complete list of metadata
Contributor : Martine Cadot <>
Submitted on : Friday, February 26, 2010 - 6:45:40 PM
Last modification on : Friday, April 2, 2021 - 3:33:32 AM


  • HAL Id : hal-00460307, version 1



Martine Cadot, Alain Lelu. Optimized Representation for Classifying Qualitative Data. Second International Conference on Advances in Databases, Knowledge, and Data Applications - DBKDA 2010, Apr 2010, Menuires, The Three Valleys, France. pp.241-246. ⟨hal-00460307⟩



Record views