Volume 19, No. 1, 2022
Detection of Depression among Social Media Users with Machine Learning
M. Senthil Raja, L. Arun Raj and A. Arun
Abstract
Mental illnesses are a significant and growing public health concern. They have the potential to tremendously affect a person’s life. Depression, in particular, is one of the major reasons for suicide. In recent times, the popularity of social media websites has burgeoned as they are platforms that facilitate discussion and free-flowing conversation about a plethora of topics. Information and dialogue about subjects like mental health, which are still considered as a taboo in various cultures, are becoming more and more accessible. The objective of this paper is to review and comprehensively compare various previously employed Natural Language Processing techniques for the purpose of classification of social media text posts as those written by depressed individuals. Furthermore, pros, cons, and evaluation metrics of these techniques, along with the challenges faced and future directions in this area of research are also summarized.
Pages: 250-257
DOI: 10.14704/WEB/V19I1/WEB19019
Keywords: Tokenization, Stemming, TF-IDF