A baseline approach for early detection of signs of anorexia and self-harm in reddit posts

Naderi, Nona (Haute école de gestion de Genève, HES-SO // Haute Ecole Spécialisée de Suisse Occidentale ; Swiss Institute of Bioinformatics, Geneva, Switzerland ; University of Toronto, Canada) ; Gobeill, Julien (Haute école de gestion de Genève, HES-SO // Haute Ecole Spécialisée de Suisse Occidentale ; Swiss Institute of Bioinformatics, Geneva, Switzerland) ; Teodoro, Douglas (Haute école de gestion de Genève, HES-SO // Haute Ecole Spécialisée de Suisse Occidentale ; Swiss Institute of Bioinformatics, Geneva, Switzerland) ; Pasche, Emilie (Haute école de gestion de Genève, HES-SO // Haute Ecole Spécialisée de Suisse Occidentale ; Swiss Institute of Bioinformatics, Geneva, Switzerland) ; Ruch, Patrick (Haute école de gestion de Genève, HES-SO // Haute Ecole Spécialisée de Suisse Occidentale ; Swiss Institute of Bioinformatics, Geneva, Switzerland)

This paper describes the systems developed by the BiTeM team for the CLEF eRisk Task 1 and 2, 2019. The goal was to predict the risk of anorexia and self-harm from user-generated content on Reddit. Several approaches based on supervised learning were used to estimate the risk of anorexia and self-harm. The systems were able to achieve low to moderate results.


Keywords:
Conference Type:
full paper
Faculty:
Economie et Services
School:
HEG - Genève
Institute:
CRAG - Centre de Recherche Appliquée en Gestion
Subject(s):
Sciences de l'information
Publisher:
Lugano, Switzerland, 9-12 September 2019
Date:
2019-09
Lugano, Switzerland
9-12 September 2019
Pagination:
12 p.
Published in:
Proceedings of CLEF (Conference and Labs of the Evaluation Forum) 2019 Working Notes
External resources:
Appears in Collection:



 Record created 2019-10-03, last modified 2019-10-04

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)