An Ensemble Learning Approach for the Detection of Depression and Mental Illness over Twitter Data

Depression and mental illness are becoming an indispensable concern, primarily among the youth. According to doctors, about 80 to 90 percent of people with depression eventually respond well to treatment. The close correspondence between social media platforms and their users helps in getting insight into the users' personal life on many levels. This project aims to analyze the tweets for self-assessed depressive features, which can make it possible for individuals, parents, caregivers, and medical professionals to combat this disorder. The project helps to identify the linguistic features of the tweets and the behavioral pattern of the Twitter users who post them, which could demonstrate symptoms of depression. This can be considered as an enhancement in the health care industry providing aid in the early detection and treatment of depression. Our proposed model works by synchronizing different machine learning algorithms to work as an ensemble model for higher efficiency and accuracy.