Journal article · 2017 · Open access

Predicting Depression: a comparative study of machine learning approaches based on language usage

Luciana Mariñelarena Dondena, Edgardo Ferretti, Manolis Μaragoudakis, Maximiliano Sapino, Marcelo Luis Errecalde

Cuadernos de neuropsicología · Published 1 January 2017 · DOI 10.7714/cnps/11.3.201

Free PDFView at the publisherCite this paper

In brief

“Predicting Depression: a comparative study of machine learning approaches based on language usage” is a 2017 journal article by Luciana Mariñelarena Dondena and 4 co-authors, published in Cuadernos de neuropsicología. OpenAlex counts 9 citations as of 15 November 2026. It is open access under a CC BY-NC-SA licence: the full text is free to read at Cuadernos de neuropsicología.

Citations
9
Most cited in
2019 (3)
Field-weighted impact
0.94
References
4
Authors
5

Abstract

Depression is a major public health concern and a leading cause of disability. Globally, more than 332 million people of all ages suffer from depression. Several studies in the literature show that people language usage is indicative of their psychological states. That is why, there is growing interest in the application of natural language processing techniques for predicting depression. In this work, we present a comparative study of different machine learning methods and alternative ways of representing documents to automatically detect social media users who have reported to had been diagnosed with depression. The obtained results have demonstrated that a Deep Learning approach had the superior classification performance, when combined with a Synthetic Minority Oversampling Technique to deal with the problem of class imbalances in the dataset used in our experiments. The F1 score achieved was 82.93% with an accuracy of more than 94%.Keywords: Depression; Machine Learning; Deep Learning; SMOTE (Synthetic Minority Oversampling TEchnique). ResumenLa depresión es uno de los mayores problemas de salud pública que constituye a su vez una de las principales causas de incapacidad. A nivel mundial, más de 332 millones de personas de todas las edades padecen este trastorno. Investigaciones previas demuestran que el lenguaje que utilizan las personas refleja su salud mental. Por tal motivo, existe un creciente interés en la aplicación de técnicas de procesamiento del lenguaje natural para predecir la depresión. En este trabajo se presenta un estudio comparativo de diferentes métodos de aprendizaje automático como así también distintas maneras de representación de los documentos con el fin de detectar automáticamente a aquellos usuarios de medios sociales que manifestaron haber sido diagnosticados previamente con depresión. Los resultados obtenidos mostraron que la performance del clasificador mejoró considerablemente cuando se aplicó un enfoque de Aprendizaje Profundo combinándolo con el algoritmo SMOTE (Synthetic Minority Oversampling TEchnique) que permite hacer frente al problema de las clases desbalanceadas alcanzando así una medida F1 del 82.93%. En síntesis, este enfoque combinado SMOTE + Aprendizaje Profundo predice la depresión con una exactitud de más del 94%.Palabras Clave: Depresión; Aprendizaje Automático; Aprendizaje Profundo; SMOTE (Synthetic Minority Oversampling TEchnique). ResumoA depressão é um dos maiores problemas de saúde pública que constitui por sua vez uma das principais causas de incapacidade. A nível mundial, mais de 332 milhões de pessoas de todas as idades padecem deste transtorno. Investigações prévias demonstram que a linguagem que utilizam as pessoas refletem a sua saúde mental. Por tal motivo, existe um crescente interesse na aplicação de técnicas de processamento da linguagem natural para prognosticar a depressão. Neste…

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Dondena, L. M., Ferretti, E., Μaragoudakis, M., Sapino, M., & Errecalde, M. L. (2017). Predicting Depression: a comparative study of machine learning approaches based on language usage. Cuadernos de neuropsicología. https://doi.org/10.7714/cnps/11.3.201

MLA 9

Dondena, Luciana Mariñelarena, et al. “Predicting Depression: a comparative study of machine learning approaches based on language usage.” Cuadernos de neuropsicología, 2017. https://doi.org/10.7714/cnps/11.3.201.

Chicago

Dondena, Luciana Mariñelarena, Edgardo Ferretti, Manolis Μaragoudakis, Maximiliano Sapino, and Marcelo Luis Errecalde. 2017. “Predicting Depression: a comparative study of machine learning approaches based on language usage.” Cuadernos de neuropsicología. https://doi.org/10.7714/cnps/11.3.201.

Harvard

Dondena, L. M. et al. (2017) ‘Predicting Depression: a comparative study of machine learning approaches based on language usage’, Cuadernos de neuropsicología. doi:10.7714/cnps/11.3.201.

IEEE

L. M. Dondena, E. Ferretti, M. Μaragoudakis, M. Sapino, and M. L. Errecalde, “Predicting Depression: a comparative study of machine learning approaches based on language usage,” Cuadernos de neuropsicología, 2017, doi: 10.7714/cnps/11.3.201.

BibTeX

@article{dondena2017predicting,
  title     = {{Predicting Depression: a comparative study of machine learning approaches based on language usage}},
  author    = {Dondena, Luciana Mariñelarena and Ferretti, Edgardo and Μaragoudakis, Manolis and Sapino, Maximiliano and Errecalde, Marcelo Luis},
  journal   = {Cuadernos de neuropsicología},
  year      = {2017},
  doi       = {10.7714/cnps/11.3.201},
  url       = {https://doi.org/10.7714/cnps/11.3.201}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
201933
202122
202211
202311
202411
202511

Its field-weighted citation impact is 0.94: it has been cited 0.94 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Social Sciences › Psychology › Social Psychology › Mental Health via Writing

  • depression prediction
  • social media text
  • language use
  • mental health detection
  • deep learning classification
  • SMOTE
  • class imbalance
  • document representation

Work with this paper

Most-cited papers in Psychology

All highly cited Psychology papers

Questions about this paper

How do I cite this paper in APA style?

Dondena, L. M., Ferretti, E., Μaragoudakis, M., Sapino, M., & Errecalde, M. L. (2017). Predicting Depression: a comparative study of machine learning approaches based on language usage. Cuadernos de neuropsicología. https://doi.org/10.7714/cnps/11.3.201

How many times has this paper been cited?

9 times, according to OpenAlex on 15 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: read the free full text at Cuadernos de neuropsicología.

Who wrote this paper?

Luciana Mariñelarena Dondena, Edgardo Ferretti, Manolis Μaragoudakis, Maximiliano Sapino, Marcelo Luis Errecalde.

What is the DOI of this paper?

10.7714/cnps/11.3.201. The permanent link is https://doi.org/10.7714/cnps/11.3.201.

Bibliographic data from OpenAlex (CC0), last checked 15 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.