Chatbots are computer programs that simulate human conversation using machine learning and artificial intelligence. Chatbots have been applied to numerous fields, including customer service, education, and physical and mental health. In the context of mental health, chatbots provide an alternative care solution in the face of barriers such as insufficient supply for the current demand, cost, and lack of convenience. Mental health chatbots exist in abundance -- around 1000 are available on the App Store today. While many are meritorious, there are a common set of critiques. Existing mental health AI agents often feel scripted, unvaried, and unable to respond to the unique, nuanced concerns of users. Overall, there is a lack of personalization -- a characteristic that is still unique to person-to-person therapy. This research aims to improve personalization of chatbots with the addition of contextual bandits, in which different contexts are provided to adaptively improve the care provided to individual users.
forked from Intelligent-Adaptive-Interventions-Lab/acai
-
Notifications
You must be signed in to change notification settings - Fork 0
Intelligent-Adaptive-Interventions-Lab/acai-simulation
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Personalized mental health chatbot using GPT-3 (Used for bot-to-bot simulation)
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published
Languages
- JavaScript 71.2%
- HTML 13.9%
- Python 12.7%
- CSS 2.2%