On Evaluating Interestingness Measures for Closed Itemsets

Aleksey Buzmakov 1, 2 Sergei O. Kuznetsov 2 Amedeo Napoli 1
1 ORPAILLEUR - Knowledge representation, reasonning
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : There are a lot of measures for selecting interesting itemsets. But which one is better? In this paper we introduce a methodology for evaluating interesting-ness measures. This methodology relies on supervised classification. It allows us to avoid experts and artificial datasets in the evaluation process. We apply our method-ology to evaluate promising measures for itemset selection, such as leverage and stability. We show that although there is no evident winner between them, stability has a slightly better performance.
Type de document :
Communication dans un congrès
Ulle Endriss, João Leite. 7th European Starting AI Researcher Symposium (STAIRS 2014), 2014, Prague, Czech Republic. 264, pp.71 - 80, 2014, Proceedings of the 7th European Starting AI Researcher Symposium (STAIRS 2014). 〈10.3233/978-1-61499-421-3-71〉
Liste complète des métadonnées

Littérature citée [17 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01095927
Contributeur : Aleksey Buzmakov <>
Soumis le : mardi 16 décembre 2014 - 14:56:02
Dernière modification le : jeudi 11 janvier 2018 - 06:25:24
Document(s) archivé(s) le : lundi 23 mars 2015 - 14:06:55

Fichier

stairs-interestingness-measure...
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Aleksey Buzmakov, Sergei O. Kuznetsov, Amedeo Napoli. On Evaluating Interestingness Measures for Closed Itemsets. Ulle Endriss, João Leite. 7th European Starting AI Researcher Symposium (STAIRS 2014), 2014, Prague, Czech Republic. 264, pp.71 - 80, 2014, Proceedings of the 7th European Starting AI Researcher Symposium (STAIRS 2014). 〈10.3233/978-1-61499-421-3-71〉. 〈hal-01095927〉

Partager

Métriques

Consultations de la notice

224

Téléchargements de fichiers

134