Prof. Tulika Saha, Assistant Professor at IIIT-Bangalore, has been honored with the Nasscom AI Game Changer Award 2023-24 in the AI Research category. Her innovative work, “A Shoulder to Cry On: Towards a Motivational Virtual Assistant for Assuaging Mental Agony,” which is a part of her doctoral thesis exemplifies the transformative potential of AI in revolutionizing mental health support. Some of Prof. Tulika’s research focuses on developing AI-driven solutions to provide empathetic, motivational guidance to individuals in mental distress. In her own words, she shares the journey behind this groundbreaking project and its far-reaching impact.
“I am deeply honored to receive the Nasscom AI Game Changer Award 2023-24 in the AI Research category for my work on “A Shoulder to Cry On: Towards a Motivational Virtual Assistant for Assuaging Mental Agony.” I congratulate and thank all the co-authors of this work which was accepted at NAACL, 2022. This award is a testament to the potential of AI in addressing one of the most pressing global health issues: mental health disorders.
The Problem We Sought to Solve
Mental health disorders, such as depression, anxiety, PTSD etc., affect millions globally. Yet, access to professional mental health care remains severely limited due to stigma, lack of awareness, and a shortage of qualified professionals. This leads many people to seek solace in online peer-to-peer platforms, where they can anonymously share their struggles and seek support from others. However, the well-intentioned but often untrained support provided by peers is not always effective, leaving many without the help they truly need.
The AI Solution: Motivational Virtual Assistant – MotiVAte
In response to this need, my team and I developed a Motivational Virtual Assistant (VA), named MotiVAte, which can provide motivational and empathetic support to individuals seeking help for mental health issues. Unlike existing conversational agents, MotiVAte is specifically designed to assuage mental agony by offering responses that are not only empathetic but also tailored to the individual’s mental health condition. The VA acts as the first point of contact, providing comfort and guidance to users suffering from various mental disorders like Major Depressive Disorder (MDD), Anxiety, Obsessive-Compulsive Disorder (OCD), and Post-Traumatic Stress Disorder (PTSD).
How It Works
At the core of MotiVAte is a two-part system:
- Mental Illness Classification (MIC): This uses an attention-based BERT classifier to identify the user’s mental health condition from the ongoing conversation.
- Motivational Dialogue Generation (MI-MDG): This module generates sentiment-driven, motivational responses based on the identified mental condition. It leverages Reinforcement Learning (RL) to optimize responses that uplift the user’s mood and encourage positive thinking.
Our VA system analyzes conversations, classifies the user’s mental health condition, and generates motivational responses that are specifically tailored to the user’s needs, providing a form of empathy at scale.
The Significance and Future Impact
The significance of this research lies in its potential to fill a critical gap in mental health support. By providing an always-available, empathetic virtual assistant, we can offer immediate relief to individuals in distress. This system doesn’t replace human mental health professionals but acts as a supplementary tool, offering motivational support while individuals seek professional help.
The potential applications of MotiVAte extend beyond mental health care. Its underlying technology can be adapted for use in customer service, education, and even elder care, where personalized, empathetic dialogue is essential.
This award reflects the growing importance of AI in healthcare and its potential to transform lives. As I continue this journey, I am excited to explore how AI can further support mental health and other domains where empathy and motivation are key to human well-being. This work has been a milestone, but it’s just the beginning of using AI for compassionate, impactful support.”
-Prof. Tulika Saha, Assistant Professor, IIIT-Bangalore
