Depression Detection Using Machine Learning Techniques On Twitter Data

Abstract

Depression has become a significant issue in today’s
generation, with the number of individuals affected
by depression increasing daily. While some recognize
they are dealing with depression, others remain
unaware. Additionally, the rise of social media has
created a new platform where users express their
mental states, often serving as a "diary." Research
has explored the use of machine learning algorithms
to identify depression through users’ social media
posts. These algorithms can classify data into
depressive and non-depressive categories. This
research aims to detect user depression based on
their social media content, specifically Twitter posts,
using two classifiers: Naïve Bayes and a hybrid
model, NBTree. The effectiveness of these classifiers
is compared based on accuracy to identify the best
method for depression detection. The results indicate
that both algorithms perform equally well, showing
the same level of accuracy [1,2].