Abstract
Depression, a significant psychiatric disorder, affects individuals' physical well-being and daily functioning. This focused analysis provides a comprehensive exploration of contemporary research conducted between 2012 and 2023 that delves into the utilization of sophisticated machine learning methodologies aimed at identifying correlates of depression within social media content. Our study meticulously dissects various data sources and performs a comprehensive examination of different machine learning algorithms cited in the researched articles and literature, aiming to pinpoint an approach that can enhance detection accuracy. Furthermore, we have scrutinized the use of varied data from social media platforms and pinpointed emerging trends, notably spotlighting novel applications of artificial neural networks for image processing and classification, along with advanced gait image models. Our results offer essential direction for future research focused on enhancing detection precision, acting as a valuable reference for academic and industry scholars in this field.
| Original language | English |
|---|---|
| Pages (from-to) | 1-28 |
| Number of pages | 28 |
| Journal | International Journal on Semantic Web and Information Systems |
| Volume | 20 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - Mar 2024 |
Scopus Subject Areas
- Information Systems
- Computer Networks and Communications
User-Defined Keywords
- Depression Detection
- Machine Learning
- Multimodal
- Sentiment Analysis
- Social Media
- Text Mining
