Constraint Acquisition Using Recommendation Queries


Abstract : Constraint acquisition systems assist the non-expert user in modeling her problem as a constraint net- work. Most existing constraint acquisition systems interact with the user by asking her to classify an example as positive or negative. Such queries do not use the structure of the problem and can thus lead the user to answer a large number of queries. In this paper, we propose PREDICT&ASK, an algo- rithm based on the prediction of missing constraints in the partial network learned so far. Such missing constraints are directly asked to the user through recommendation queries, a new, more informative kind of queries. PREDICT&ASK can be plugged in any constraint acquisition system. We experimen- tally compare the QUACQ system to an extended version boosted by the use of our recommendation queries. The results show that the extended version improves the basic QUACQ.
Type de document :
Communication dans un congrès
IJCAI: International Joint Conference on Artificial Intelligence, Jul 2016, New York City, United States. 25th International Joint Conference on Artificial Intelligence, pp.720-726, 2016, 〈http://ijcai-16.org/〉
Liste complète des métadonnées

Littérature citée [12 références]  Voir  Masquer  Télécharger

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01374716
Contributeur : Joël Quinqueton <>
Soumis le : vendredi 30 septembre 2016 - 21:31:51
Dernière modification le : dimanche 16 décembre 2018 - 10:42:02
Document(s) archivé(s) le : samedi 31 décembre 2016 - 16:33:17

Fichier

ijcai16-recommendation.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : lirmm-01374716, version 1

Collections

Citation

Abderrazak Daoudi, Younes Mechqrane, Christian Bessière, Nadjib Lazaar, El Houssine Bouyakhf. Constraint Acquisition Using Recommendation Queries
. IJCAI: International Joint Conference on Artificial Intelligence, Jul 2016, New York City, United States. 25th International Joint Conference on Artificial Intelligence, pp.720-726, 2016, 〈http://ijcai-16.org/〉. 〈lirmm-01374716〉

Partager

Métriques

Consultations de la notice

132

Téléchargements de fichiers

156