Answers are grounded in this paper and cite their sources.
Depressive disorder is a psychiatric problem that has a high contribution to the cause of disability in the world, as it can cause people who experience it to not perform well in daily life which might affect their work life.To help patients who needed immediate treatment, the World Health Organization (WHO) has created a program called Mental Health Gap Action Program (mhGAP) in order to increase development that focuses on monitoring mental health in a globalized world.This research explores several different classification model architectures with deep learning approaches in order to detect Twitter users' tweets that were classified as tweets with depressive disorder and then compare the results between the proposed models.As a result, the Long Short-Term Memory (LSTM) model paired with Transformer word representation models has provided better performance when compared to employing Bidirectional Encoder Representations from Transformers (BERT) baseline classification or LSTM classification models alone.The evaluation of the classification model's architecture reveals that the LSTM model, when paired with MentalBERT, exhibits the highest accuracy performance, achieving an accuracy score of 0.86.This approach leverages the respective advantages of transformer and LSTM models to facilitate the accurate analysis of syntactic and contextual information associated with individual words, hence enabling the appropriate representation or domain-specific of sequential data.
No takes yet. Share an insight, caveat, or question.
Anindyaputri et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: