Citation

Kwee HR, Zahra A. ICIC Express Letters, Part B: Applications 2022; 13(9): 983-989.

Copyright

(Copyright © 2022)

DOI

10.24507/icicelb.13.09.983

PMID

unavailable

Abstract

Depression is one of the mental disorders suffered by many people in the world. People usually keep it to themselves and do not seek treatment. Proper treatment is needed to prevent people from suicide. To know how people with depression behave, we can diagnose them from the language that they used every day. Because the Internet is one of the most used services in the world, we can learn from the text they posted on social media. In this work, we used Reddit posts as the main data source to learn about depressive text. We then finetuned BERT and DeBERTa models with the collected dataset. Using this method, the highest accuracy achieved is 96.76% using the BERT model. From our experimental results, BERT and DeBERTa are able to learn about depressive and non-depressive text. © 2022, ICIC International.

Language: en

Keywords

BERT; DeBERTa; Depression; Depression detection; Reddit; Transfer learning; Transformer model