Baymaxxing

Inspiration

Mental health is an ongoing challenge, and many people struggle to find immediate support or resources tailored to their experiences. Inspired by Baymax from Big Hero 6, we wanted to create an AI-powered companion that analyzes emotions, offers support, and connects users with relevant real-world discussions—all in a seamless and accessible way.

What it does

Baymaxxing acts as a personal AI mental wellness assistant by:

Accepting user-inputted text to analyze sentiment.
Categorizing the text as anxiety, depression, or mental illness with a severity score (1-10) using Groq AI.
Providing AI-generated words of support based on the severity score.
Displaying top 3 Reddit posts with similar sentiment and severity for users to find relatable discussions.
Offering data-driven insights using Tableau to visualize trends in mental health sentiment.

How we built it

Data Processing: A Python Jupyter Notebook script scrapes Reddit posts, runs them through Groq AI to categorize and give a severity rating for each reddit post, and stores results in a CSV file. Backend: FastAPI handles API requests, processes sentiment analysis, and fetches Reddit data. Frontend: Built with Next.js and Tailwind CSS to create a smooth, intuitive user experience. Visualization: Tableau is used to analyze and present insights from the dataset.

Challenges we ran into

Reddit scraping limitations: Ensuring we collected meaningful posts while handling API rate limits.
Data Visualization on frontend: After we used the collected data to create a dashboard using Tableau, we had trouble displaying Tableau on our nextjs frontend.
Displaying Reddit Post based on category and severity: Had trouble narrowing down the reddit posts based on the category and severity from the frontend.

Accomplishments that we're proud of

Successfully integrating AI-powered sentiment analysis into a functional web app.
Building a fully working prototype within a limited timeframe.
Creating a meaningful project that promotes mental health awareness.
Implementing real-time user interaction with AI-generated support.

What we learned

Improved our understanding of FastAPI, Next.js, and AI-driven sentiment analysis.
Learned how to scrape and process large datasets efficiently.
Understood how to balance data-driven insights with an empathetic user experience.
Gained experience in full-stack development and cloud-based AI models.

What's next for Baymaxxing

Enhancing AI responses to provide more personalized and empathetic support.
Expanding dataset sources beyond Reddit for more diverse insights.
Improving severity classification using additional AI training.
Deploying the app for broader accessibility.
Adding real-time support features, such as chatbot interactions or mental health resources.