Mining Tweet Data - Statistic and semantic information for political tweet classification

Guillaume Tisserant 1 Violaine Prince 1 Mathieu Roche 2, 3
1 TEXTE - Exploration et exploitation de données textuelles
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
2 ADVANSE - ADVanced Analytics for data SciencE
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
3 UMR TETIS - Territoires, Environnement, Télédétection et Information Spatiale
Abstract : This paper deals with the quality of textual features in messages in order to classify tweets. The aim of our study is to show how improving the representation of textual data affects the performance of learning algorithms. We will first introduce our method GenDesc. It generalises less relevant words for tweet classi- fication. Secondly we compare and discuss the types of textual features given by different approaches. More precisely we discuss the semantic specificity of textual features, e.g. Named Entity, HashTag.
Type de document :
Communication dans un congrès
KDIR: Knowledge Discovery and Information Retrieval, Oct 2014, Rome, Italy. 6th International Conference on Knowledge Discovery and Information Retrieval, pp.523-529, 2014, Text-Mining Session. 〈http://www.kdir.ic3k.org/?y=2014〉. 〈10.5220/0005170205230529〉
Liste complète des métadonnées
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01054908
Contributeur : Mathieu Roche <>
Soumis le : samedi 9 août 2014 - 21:10:48
Dernière modification le : mardi 18 décembre 2018 - 12:20:54

Identifiants

Collections

CIRAD | AGROPARISTECH | IRSTEA | TEXTE | AGROPOLIS | LIRMM | ADVANSE | MIPS | UNIV-MONTPELLIER

Citation

Guillaume Tisserant, Violaine Prince, Mathieu Roche. Mining Tweet Data - Statistic and semantic information for political tweet classification. KDIR: Knowledge Discovery and Information Retrieval, Oct 2014, Rome, Italy. 6th International Conference on Knowledge Discovery and Information Retrieval, pp.523-529, 2014, Text-Mining Session. 〈http://www.kdir.ic3k.org/?y=2014〉. 〈10.5220/0005170205230529〉. 〈lirmm-01054908〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

247

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD