| Titre : | A Novel Semi-Supervised Approach for Depression Detection and Monitoring Over Time | | Type de document : | projet fin études | | Auteurs : | Ayoub RMIDI, Auteur | | Langues : | Français (fre) | | Catégories : | Ingénierie Finance et la Gestion des Risques (IFGR)
| | Mots-clés : | Semi-supervised Machine Learning, Natural Language Processing,
Social Media, Mental Health | | Index. décimale : | mast 142/18 | | Résumé : | With the rapid development of social media sites, millions of people share their
thoughts and feelings on platforms such as Twitter, Reddit and Facebook. These
platforms provide a new way to detect mental health illness evolution like depression
and suicidal ideation. Such a task is in fact still very challenging for many reasons;
one of which is the lack of labeled dataset by domain experts. Researchers have tried
before to study depression from a single sentence as a typical classification problem,
which is not enough to tell us about depression existence. Unlike the proposed
state of the art supervised and deep learning techniques, our approach does not
require any labeled dataset. In this research, we propose a novel semi-supervised
approach to detect and monitor depression symptoms over time through analysis
of tweets posted by individuals. This approach combines lexical, topic modeling
and word embedding-based approach. Our experiments show promising results in
terms of topic coherency, which motivate us to investigate a symptoms multi-label
classification problem.
|
A Novel Semi-Supervised Approach for Depression Detection and Monitoring Over Time [projet fin études] / Ayoub RMIDI, Auteur . - [s.d.]. Langues : Français ( fre) | Catégories : | Ingénierie Finance et la Gestion des Risques (IFGR)
| | Mots-clés : | Semi-supervised Machine Learning, Natural Language Processing,
Social Media, Mental Health | | Index. décimale : | mast 142/18 | | Résumé : | With the rapid development of social media sites, millions of people share their
thoughts and feelings on platforms such as Twitter, Reddit and Facebook. These
platforms provide a new way to detect mental health illness evolution like depression
and suicidal ideation. Such a task is in fact still very challenging for many reasons;
one of which is the lack of labeled dataset by domain experts. Researchers have tried
before to study depression from a single sentence as a typical classification problem,
which is not enough to tell us about depression existence. Unlike the proposed
state of the art supervised and deep learning techniques, our approach does not
require any labeled dataset. In this research, we propose a novel semi-supervised
approach to detect and monitor depression symptoms over time through analysis
of tweets posted by individuals. This approach combines lexical, topic modeling
and word embedding-based approach. Our experiments show promising results in
terms of topic coherency, which motivate us to investigate a symptoms multi-label
classification problem.
|
|
Faire une suggestion Affiner la recherche
A Novel Semi-Supervised Approach for Depression Detection and Monitoring Over Time / Ayoub RMIDI
