Temporal Mood Variation: at the CLEF eRisk-2018 Tasks for Early Risk Detection on The Internet

Waleed Ragheb 1 Bilel Moulahi 1 Jérôme Azé 1 Sandra Bringay 1 Maximilien Servajean 1
1 ADVANSE - ADVanced Analytics for data SciencE
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Two tasks are proposed at CLEF eRisk-2018 on predicting mental disorder using Users posts on Reddit. Depression and anorexia disorders are considered to be detected as early as possible. In this paper we present the participation of LIRMM (Laboratoire d’Informatique, de Robotique et de Micro´electronique de Montpellier) in both tasks. The proposed architectures and models use only text information without any hand-crafted features or dictionaries to model the temporal mood variation detected from users posts. The proposed models use two learning phases through exploration of state-of-the-art text vectorization. The proposed models perform comparably to other contributions while experiments shows that document-level outperformed word-level vectorizations.
Keywords : Classification Word2vec Doc2vec Temporal variation MLP Depression Anorexia
Document type :
Conference papers
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01989632
Contributor : Waleed Ragheb <>
Submitted on : Tuesday, January 22, 2019 - 2:53:10 PM
Last modification on : Thursday, February 7, 2019 - 3:41:57 PM
Long-term archiving on : Tuesday, April 23, 2019 - 3:22:40 PM

File

paper_78.pdf

Identifiers

  • HAL Id : lirmm-01989632, version 1

Collections

LIRMM | ADVANSE | MIPS | UNIV-MONTPELLIER

Citation

Waleed Ragheb, Bilel Moulahi, Jérôme Azé, Sandra Bringay, Maximilien Servajean. Temporal Mood Variation: at the CLEF eRisk-2018 Tasks for Early Risk Detection on The Internet. CLEF: Conference and Labs of the Evaluation, Sep 2018, Avignon, France. pp.#78. ⟨lirmm-01989632⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

66

Files downloads

60

Contact

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