Citation

Narynov S, Mukhtarkhanuly D, Omarov B. Data Brief 2020; 29: e105195.

Affiliation

Khoja Akhmet Yassawi International Kazakh-Turkish University, Turkistan, Kazakhstan.

Copyright

(Copyright © 2020, Elsevier Publishing)

DOI

10.1016/j.dib.2020.105195

PMID

32083154

PMCID

PMC7016367

Abstract

This paper presents dataset collected from social networks that are mostly used by youth of Commonwealth of Independent States (CIS) countries. The data was collected from public accounts of VKontakte social network by using VK.api and applying the most used keywords that would signify depressive mood. The collected data was classified by psychologists into two types: depressive and non-depressive. The dataset consists of 32 018 depressive posts and 32 021 non-depressive posts. Since the most common language that is spoken in CIS countries is Russian, the posts are written in Russian, consequently the collected data is in Russian language as well. The data can mostly be useful for researchers who explore tendencies to depression in CIS countries. The dataset is important for the research community, as it was not only collected from open sources, but also marked by our psychiatrists from the republican scientific and practical center of mental health. Since the dataset has very high validity, it can be used for further research in the field of mental health.

© 2020 The Author(s).

Language: en

Keywords

Dataset; Depression; Machine learning; NLP; Social network; Suicide