EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Depression in South Korean Adolescents Captured by Text and Opinion Mining of Social Big Data

Juyoung Song iD, Tae-Min Song, Sangho Lee iD, Dong-Chul Seo iD

International Journal of Environmental Research and Public Health · 2023

DOI: 10.3390/ijerph20176665 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Environmental Research and Public Health, 2023. Verfasst von Juyoung Song und 3 weiteren Autor:innen. Beteiligte Einrichtung: Administration of Justice, Pennsylvania State University, Schuylkill, PA 17972, USA. Hauptthemen laut MEDLINE: Sentiment Analysis, Big Data. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Depression in adolescence is recognized as an important social and public health issue that interferes with continued physical growth and increases the likelihood of other mental disorders. The goal of this study was to examine online documents posted by South Korean adolescents for 3 years through the text and opinion mining of collectable documents in order to capture their depression. The sample for this study was online text-based individual documents that contained depression-related words among adolescents, and these were collected from 215 social media websites in South Korea from 1 January 2012 to 31 December 2014. A sentiment lexicon was developed for adolescent depressive symptoms, and such sentiments were analyzed through opinion mining. The depressive symptoms in the present study were classified into nine categories as suggested by the Diagnostic and Statistical Manual for Mental Disorders, 5th Edition (DSM-5). The association analysis and decision tree analysis of data mining were used to build an efficient prediction model of adolescent depression. Opinion mining indicated that 15.5% were emotionally stable, 58.6% moderately stressed, and 25.9% highly distressed. Data mining revealed that the presence of depressed mood most of the day or nearly every day had the greatest effect on adolescents’ depression. Social big data analysis may serve as a viable option for developing a timely response system for emotionally susceptible adolescents. The present study represents one of the first attempts to investigate depression in South Korean adolescents using text and opinion mining from three years of online documents that originally amounted to approximately 3.1 billion documents.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Sentiment AnalysisBig DataAdolescentHumansDepressionData MiningRepublic of Korea

Stichwörter der Autor:innen

adolescentsdepressionemotional susceptibilitysocial big datatext mining

Autor:innen und Institutionen

  • Administration of Justice, Pennsylvania State University, Schuylkill, PA 17972, USA.
  • Gachon University Graduate School of Industry & Environment, Seoul 13120, Republic of Korea.
  • HealthMax Co., Ltd., Seoul 06078, Republic of Korea.
  • Department of Applied Health Science, Indiana University School of Public Health, Bloomington, IN 47405, USA.

ORCID-Profile: Juyoung Song, Sangho Lee, Dong-Chul Seo

Zitieren

APA

Song, J., Song, T. M., Lee, S., & Seo, D. C. (2023). Depression in South Korean Adolescents Captured by Text and Opinion Mining of Social Big Data. International Journal of Environmental Research and Public Health. https://doi.org/10.3390/ijerph20176665

Vancouver

Song J, Song TM, Lee S, Seo DC. Depression in South Korean Adolescents Captured by Text and Opinion Mining of Social Big Data. International Journal of Environmental Research and Public Health. 2023. doi:10.3390/ijerph20176665

BibTeX
@article{Song2023,
  author = {Song, Juyoung and Song, Tae-Min and Lee, Sangho and Seo, Dong-Chul},
  title = {Depression in South Korean Adolescents Captured by Text and Opinion Mining of Social Big Data},
  journal = {International Journal of Environmental Research and Public Health},
  year = {2023},
  doi = {10.3390/ijerph20176665}
}
Kostenlos starten →