Feature Construction and δ-Free Sets in 0/1 Samples

Abstract : Given the recent breakthrough in constraint-based mining of local patterns, we decided to investigate its impact on feature construction for classification tasks. We discuss preliminary results concerning the use of the so-called δ-free sets. Our guess is that their minimality might help to collect important features. Once these sets are computed, we propose to select the essential ones w.r.t. class separation and generalization as new features. Our experiments have given encouraging results.
Document type :
Conference papers
Liste complète des métadonnées

http://hal.univ-reunion.fr/hal-01464502
Contributor : Réunion Univ <>
Submitted on : Friday, February 10, 2017 - 11:43:23 AM
Last modification on : Wednesday, April 3, 2019 - 1:05:20 AM

Identifiers

  • HAL Id : hal-01464502, version 1

Citation

Nazha Selmaoui-Folcher, Claire Leschi, Dominique Gay, Jean-François Boulicaut. Feature Construction and δ-Free Sets in 0/1 Samples. Discovery Science, 9th International Conference, DS 2006, Oct 2016, Barcelona, Spain. pp.363-367. ⟨hal-01464502⟩

Share

Metrics

Record views

243