Towards Dynamic Classification Completeness in Twitter

Abstract : In this paper we study the application of Matrix Completion in topic detection and classification in Twitter. The proposed method first employs Joint Complexity to perform topic detection based on score matrices. Based on the spatial correlation of tweets and the spatial characteristics of the score matrices, we apply a novel framework which extends the Matrix Completion to build dynamically complete matrices from a small number of random sample Joint Complexity scores. The experimental evaluation with real data from Twitter presents the topic detection accuracy based on complete reconstructed matrices, and thus reducing the exhaustive computation of Joint Complexity scores.
Type de document :
Communication dans un congrès
24th IEEE European Signal Processing Conference (EUSIPCO 2016), Aug 2016, Budapest, Hungary. 2016 24th European Signal Processing Conference (EUSIPCO), 〈10.1109/EUSIPCO.2016.7760418〉
Liste complète des métadonnées

https://hal-polytechnique.archives-ouvertes.fr/hal-01340581
Contributeur : Dimitrios Milioris <>
Soumis le : jeudi 16 mars 2017 - 12:03:30
Dernière modification le : mercredi 22 mars 2017 - 01:10:27
Document(s) archivé(s) le : samedi 17 juin 2017 - 12:15:05

Fichier

Milioris_Eusipco2016.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Dimitrios Milioris. Towards Dynamic Classification Completeness in Twitter. 24th IEEE European Signal Processing Conference (EUSIPCO 2016), Aug 2016, Budapest, Hungary. 2016 24th European Signal Processing Conference (EUSIPCO), 〈10.1109/EUSIPCO.2016.7760418〉. 〈hal-01340581〉

Partager

Métriques

Consultations de la notice

112

Téléchargements de fichiers

135