期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2020
卷号：98
期号：1
页码：124-135
出版社：Journal of Theoretical and Applied
摘要：In today’s digital age, social media have become the most common channel for individuals to express their opinions and feelings. As common as this, the extensive usage of social media has also been associated with mental illnesses such as anxiety, suicidality and depression. The digital traces the individuals left provide insights into not just their daily life but also on their health and mental state. This allows for various prediction and preliminary diagnosis to be made. The advancement of research in the Natural Language Processing (NLP) field has allowed researchers to understand individuals based on texts they shared in their social media account. This paper reviews the techniques and methods used in detecting depression from social media texts where emphasis are being placed on the writing style and the word usage of the social media users. Writing styles and choices of words have been seen as a possible indicator in detecting depression from social media texts. Various methods and platforms have been adopted to investigate the effectiveness of detecting depression based on these two components. This paper discusses these methods and techniques as well as the areas where improvements can be made.
