Identifying Mental Health Conditions from Social Media using Deep Learning
Disciplines
Artificial Intelligence and Robotics | Other Computer Sciences
Abstract (300 words maximum)
Mental health issues are a continually growing concern within the youth and society today. With the age of technology, the presence of depression has become more prevalent among individuals. Social media can present a platform for one to express these emotions, exposing the presence of mental health issues like depression online. However, the detection of depression and other mental health issues from social media data is very challenging since the data amount is massive and the symptoms of depression vary for individuals. Deep learning, a subbranch of machine learning has shown promising results in detecting patterns from complex data sources. In our study, we utilized the potential of deep neural architecture to accurately identify depression from social media data. Our studies using Long short-term memory models (LSTM) and Convolutional neural networks (CNN) architecture-based models have yielded promising results in detecting depression. Our proposed model can efficiently and accurately detect depression from social media data. Accurately detecting the presence of depression will allow us to provide timely assistance to individuals who are suffering and help them recover from their condition.
Academic department under which the project should be listed
CCSE - Computer Science
Primary Investigator (PI) Name
Md Abdullah Al Hafiz Khan
Identifying Mental Health Conditions from Social Media using Deep Learning
Mental health issues are a continually growing concern within the youth and society today. With the age of technology, the presence of depression has become more prevalent among individuals. Social media can present a platform for one to express these emotions, exposing the presence of mental health issues like depression online. However, the detection of depression and other mental health issues from social media data is very challenging since the data amount is massive and the symptoms of depression vary for individuals. Deep learning, a subbranch of machine learning has shown promising results in detecting patterns from complex data sources. In our study, we utilized the potential of deep neural architecture to accurately identify depression from social media data. Our studies using Long short-term memory models (LSTM) and Convolutional neural networks (CNN) architecture-based models have yielded promising results in detecting depression. Our proposed model can efficiently and accurately detect depression from social media data. Accurately detecting the presence of depression will allow us to provide timely assistance to individuals who are suffering and help them recover from their condition.