Saathi-An AI Campanion

Authors

  • Ankita Ramesh Upadhyay Department of Computer Engineering Shree L.R. Tiwari College of Engineering Mumbai, India
  • Amey Atul Bhavsar Department of Computer Engineering Shree L.R. Tiwari College of Engineering Mumbai, India
  • Deepanshu Kailashnath Yadav Department of Computer Engineering Shree L.R. Tiwari College of Engineering Mumbai, India
  • Pooruvi Virendra Singh Department of Computer Engineering Shree L.R. Tiwari College of Engineering Mumbai, India
  • Neelam Phadnis Department of Computer Engineering Shree L.R. Tiwari College of Engineering Mumbai, India

Keywords:

Mental health, Chatbot, NLP, Depression, Anxiety, Bipolar Disorder, CNN, Emotion Detection, Neural Networks, Logistic Regression, Disease Detection, Machine Learning, Deep Learning

Abstract

— The research paper presents a novel approach for developing a mental health chatbot (Saathi) to provide empathetic and human-like support. Using an open domain system, the model behind chatbot predicts the probabilities from message for signs of mental illness like depression, anxiety, and bipolar disorder & emotions. This paper demonstrates a generative approach to create outputs sent to the user keeping the recent history of signs of mental illness and emotions as inputs to the generative model. This results in a more appropriate output since it is dependent on the current state of the user. Saathi has the potential to improve the early detection and management of mental health disorders by providing users with a convenient and confidential way to receive help and support. Focused on adolescents, we aim to bridge the gap between people and mental health aid using this application.


Downloads

Published

2023-06-01

How to Cite

Ankita Ramesh Upadhyay, Amey Atul Bhavsar, Deepanshu Kailashnath Yadav, Pooruvi Virendra Singh, & Neelam Phadnis. (2023). Saathi-An AI Campanion. SSGM Journal of Science and Engineering, 1(1), 179–183. Retrieved from https://ssgmjournal.in/index.php/ssgm/article/view/79