Efficient Active Novel Class Detection for Data Stream Classification

Mohamed-Rafik Bouguelia 1 Yolande Belaïd 1 Abdel Belaïd 1
1 READ - Recognition of writing and analysis of documents
LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : One substantial aspect of data stream classification is the possible appearance of novel unseen classes which must be identified in order to avoid confusion with existing classes. Detecting such new classes is omitted by most existing techniques and rarely addressed in the literature. We address this issue and propose an efficient method to identify novel class emergence in a multi-class data stream. The proposed method incrementally maintains a covered feature space of existing (known) classes. An incoming data point is designated as "insider" or "outsider" depending on whether it lies inside or outside the covered space area. An insider represents a possible instance of an existing class, while an outsider may be an instance of a possible novel class. The proposed method is able to iteratively select those insiders (resp. outsiders) that are more likely to be members of a novel (resp. an existing) class, and eventually distinguish the actual novel and existing classes accurately. We show how to actively query the labels of the identified novel class instances that are most uncertain. The method also allows us to balance between the rapidity of the novelty detection and its efficiency. Experiments using real world data prove the effectiveness of our approach for both the novel class detection and classification accuracy.
Type de document :
Communication dans un congrès
ICPR - International Conference on Pattern Recognition, Aug 2014, Stockholm, Sweden. IEEE, pp.2826-2831, 2014, 〈10.1109/ICPR.2014.487〉
Liste complète des métadonnées

Littérature citée [17 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/hal-01065043
Contributeur : Yolande Belaid <>
Soumis le : mercredi 17 septembre 2014 - 16:51:58
Dernière modification le : mardi 24 avril 2018 - 13:30:30
Document(s) archivé(s) le : jeudi 18 décembre 2014 - 11:50:59

Fichier

ICPR_version_editeur.pdf
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

Collections

INRIA | UNIV-LORRAINE | LORIA-NLPKD | LORIA

Citation

Mohamed-Rafik Bouguelia, Yolande Belaïd, Abdel Belaïd. Efficient Active Novel Class Detection for Data Stream Classification. ICPR - International Conference on Pattern Recognition, Aug 2014, Stockholm, Sweden. IEEE, pp.2826-2831, 2014, 〈10.1109/ICPR.2014.487〉. 〈hal-01065043〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

388

Téléchargements de fichiers

328

Contact


archive-ouverte@inria.fr