Depression is the most common mental illness of the human body and often leads the sufferers to commit suicide; as a consequence, it is necessary to detect the depressed state of the sufferer. The emotional state of the suffering plays a vital role in depression detection. With the boom of social networking, it is possible to make decisions and predictions using the data shared by the users in the form of text, voice, or video. Depression detection aims at predicting the depression level of the sufferer via information collected from social platforms.
Traditional learning models for depression solely focus on representing inherent features of the text information from the sufferer. Analyzing the emotional state of the text along with multiple modalities helps in decision-making to detect the depression state of the sufferer. Emotional feature extraction assists in categorizing and recognizing the different emotions from the user’s information to detect the depression level. Depression detection is performed well, along with emotional feature extraction.