Depression Detection from a Social Media Dataset Using Deep Learning and NLP Techniques: A Review

In recent years, the widespread occurrence of psychological state issues, especially depression, has turned into a growing concern worldwide. The increasing reliance on online media platforms as a way of communication has caused to an influx of user-generated data, presenting a fresh chance for timely diagnosis and intervention in intellectual health conditions. Initial recognition and identification are very much important in preventing adverse outcomes and raising the standard of living of depressed individuals. However, majority of group who are struggling from depression do not seek professional help due to various barriers such as stigma, cost, or lack of access. Therefore, alternative methods of identifying and monitoring depression are needed, particularly in the period of online platform, where people often express their thinking and emotion online. This literature survey aims to review the state-of-the-art methods and applications of deep learning and NLP techniques in pointing out depression from online social interactions. By analyzing recent advancements in this field, including neural network architectures and language models, this study seeks to offer a comprehensive insight into difficulties in automatic detection of depression-related cues from textual data and to inspire further innovative work in leveraging online media for psychological well-being assessment.