Depression Detection on Social Media Using Text Mining

Depression is regarded as the major cause of disability and a leading cause of suicide. Social media has recently risen to prominence as a leading means of disseminating information online. The majority of people use the internet to express their opinions, ideologies, and personal experiences. Depression has an effect on how people communicate on social media as evidenced by the language they use. The detection of mental illness has become a major concern as a result of the massive increase in mental health awareness. In this paper, we examine Twitter users’ tweets to detect any factors that may reveal depression. For this purpose, we use Text Mining and Natural Language Processing techniques. We employed CNN + LSTM classification technique and obtained an accuracy of 92%. We have also compared our model with Logistic Regression and TF-IDF classifiers.