Emotion Recognition-Based Mental Healthcare Chat-bots: A Survey

The average human attention span has cascaded down from 12 to a mere 8 seconds. This is a direct consequence of man’s tech-saturated lifestyle. Humans have shown a changing trend of preferring quick and informative digital conversations over a time-consuming human-to-human interaction. Concurrently, a humongous increase in research being done on chat-bot technology is seen. A well-trained chat-bot capable of having a fulfilling and productive conversation has seen keen interest in users. Ever since the outbreak of the COVID-19 pandemic our lives have been forced to change drastically. Difficulty managing to adapt to the post-COVID lifestyle has raised concerns about psychological resilience to adversity. The situation calls for immediate attention to mental healthcare. In this survey, a study of the latest papers on how emotion recognition, in addition to sentiment analysis can be integrated into a chat-bot to help identify and resolve a user's mental anguish is done. This survey is aimed at finding and analysing the existing methods used to develop a self-sufficient emotion recognition-based chat-bot system that can take up the role of a therapist.