Cargando…
Understanding Depression from Psycholinguistic Patterns in Social Media Texts
The World Health Organization reports that half of all mental illnesses begin by the age of 14. Most of these cases go undetected and untreated. The expanding use of social media has the potential to leverage the early identification of mental health diseases. As data gathered via social media are a...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7148066/ http://dx.doi.org/10.1007/978-3-030-45442-5_50 |
_version_ | 1783520523888623616 |
---|---|
author | Trifan, Alina Antunes, Rui Matos, Sérgio Oliveira, Jose Luís |
author_facet | Trifan, Alina Antunes, Rui Matos, Sérgio Oliveira, Jose Luís |
author_sort | Trifan, Alina |
collection | PubMed |
description | The World Health Organization reports that half of all mental illnesses begin by the age of 14. Most of these cases go undetected and untreated. The expanding use of social media has the potential to leverage the early identification of mental health diseases. As data gathered via social media are already digital, they have the ability to power up faster automatic analysis. In this article we evaluate the impact that psycholinguistic patterns can have on a standard machine learning approach for classifying depressed users based on their writings in an online public forum. We combine psycholinguistic features in a rule-based estimator and we evaluate their impact on this classification problem, along with three other standard classifiers. Our results on the Reddit Self-reported Depression Diagnosis dataset outperform some previously reported works on the same dataset. They stand for the importance of extracting psychologically motivated features when processing social media texts with the purpose of studying mental health. |
format | Online Article Text |
id | pubmed-7148066 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
record_format | MEDLINE/PubMed |
spelling | pubmed-71480662020-04-13 Understanding Depression from Psycholinguistic Patterns in Social Media Texts Trifan, Alina Antunes, Rui Matos, Sérgio Oliveira, Jose Luís Advances in Information Retrieval Article The World Health Organization reports that half of all mental illnesses begin by the age of 14. Most of these cases go undetected and untreated. The expanding use of social media has the potential to leverage the early identification of mental health diseases. As data gathered via social media are already digital, they have the ability to power up faster automatic analysis. In this article we evaluate the impact that psycholinguistic patterns can have on a standard machine learning approach for classifying depressed users based on their writings in an online public forum. We combine psycholinguistic features in a rule-based estimator and we evaluate their impact on this classification problem, along with three other standard classifiers. Our results on the Reddit Self-reported Depression Diagnosis dataset outperform some previously reported works on the same dataset. They stand for the importance of extracting psychologically motivated features when processing social media texts with the purpose of studying mental health. 2020-03-24 /pmc/articles/PMC7148066/ http://dx.doi.org/10.1007/978-3-030-45442-5_50 Text en © Springer Nature Switzerland AG 2020 This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic. |
spellingShingle | Article Trifan, Alina Antunes, Rui Matos, Sérgio Oliveira, Jose Luís Understanding Depression from Psycholinguistic Patterns in Social Media Texts |
title | Understanding Depression from Psycholinguistic Patterns in Social Media Texts |
title_full | Understanding Depression from Psycholinguistic Patterns in Social Media Texts |
title_fullStr | Understanding Depression from Psycholinguistic Patterns in Social Media Texts |
title_full_unstemmed | Understanding Depression from Psycholinguistic Patterns in Social Media Texts |
title_short | Understanding Depression from Psycholinguistic Patterns in Social Media Texts |
title_sort | understanding depression from psycholinguistic patterns in social media texts |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7148066/ http://dx.doi.org/10.1007/978-3-030-45442-5_50 |
work_keys_str_mv | AT trifanalina understandingdepressionfrompsycholinguisticpatternsinsocialmediatexts AT antunesrui understandingdepressionfrompsycholinguisticpatternsinsocialmediatexts AT matossergio understandingdepressionfrompsycholinguisticpatternsinsocialmediatexts AT oliveirajoseluis understandingdepressionfrompsycholinguisticpatternsinsocialmediatexts |