🏆 The AI Wellness Hub

A Dual-AI Platform for Proactive Workplace Mental Health

Submission for the AITHON: 14 Days of AI for Social Good

The AI Wellness Hub is a comprehensive, multi-page application designed to foster mentally healthy workplaces. It combines a powerful predictive analytics tool for managers with a safe, informational chatbot for all employees, creating a complete ecosystem for mental wellness support.

Demo: https://youtu.be/l9HKwWM9s_0

Screenshot:

---

💡 The Problem: Moving from Intention to Action

Many companies genuinely want to support their employees' mental health, but they often lack the data-driven tools to understand their unique environment. This leads to generic, one-size-fits-all solutions that are often ineffective. The key challenge is bridging the gap between the desire to help and the ability to provide targeted, impactful support.

✨ Our Solution: The AI Wellness Hub

Our Hub tackles this problem with a two-pronged, AI-powered approach:

1. 📈 The Workplace Wellness Advisor

A sophisticated diagnostic tool designed for managers and HR leaders. It provides a multi-faceted analysis of the workplace environment, turning raw data into clear, actionable strategy.

Key Features:

• Predictive Analysis: Forecasts the likelihood of employees seeking mental health treatment using a fine-tuned XGBoost model.
• Wellness Report Card: Scores the workplace on the crucial dimensions of "Support Structures" and "Open Culture."
• AI Explainability (SHAP): Moves beyond a "black box" AI by using SHAP to reveal the exact factors driving the prediction, building trust and providing deep insights.
• Generative AI Companion (Gemini): A Google Gemini-powered advisor provides a detailed, empathetic interpretation of the results with actionable recommendations.
• AI Action Kit: Instantly generates draft emails and meeting agendas to help managers implement the AI's advice.
• Downloadable PDF Report: Creates a professional, shareable PDF summary of the complete analysis for stakeholder meetings.

2. 💬 The Mental Healthcare Chatbot

A safe and anonymous informational resource for everyone. This chatbot uses Google Gemini to provide reliable, general information about mental healthcare topics. It is strictly programmed with safety protocols to not give medical advice and to immediately provide crisis hotline numbers if a user appears to be in distress.

---

🚀 The Development Journey & Technical Deep Dive

Our process was a real-world journey of iterative development and intelligent problem-solving.

1. Initial Model (V1): We began by training several models on the OSMI Mental Health in Tech Survey dataset.
2. The Critical Insight (V2): We built a prototype and discovered a critical flaw—the model was "lazy," relying on non-actionable features like Age and family_history. While statistically accurate, it was not practically useful as an advisor.
3. The Strategic Pivot (V3): We re-engineered our entire approach, creating a focused "Advisor Model" by intentionally removing non-actionable features. This forced the AI to learn the more subtle patterns within workplace policies, making the tool genuinely responsive.
4. Advanced Data Techniques (V4): To enhance our focused model, we implemented IterativeImputer to intelligently handle missing data and SMOTE to synthetically augment our training set, leading to a more robust final model.
5. The Complete Application (V5): We built the final multi-page application, integrating all advanced features like SHAP, Gemini AI, PDF reports, and the informational chatbot.

Tech Stack

• Programming Language: Python
• Web Framework: Streamlit
• Machine Learning: Scikit-learn, XGBoost, Pandas, NumPy
• AI Explainability: SHAP
• Generative AI: Google Gemini API
• Data Augmentation: Imbalanced-learn (for SMOTE)
• PDF Generation: FPDF2
• Data Visualization: Matplotlib

---

🛠️ How to Run the Project Locally

This project is fully reproducible. Follow these steps to run it on your local machine.

Prerequisites

• Python 3.9+
• An API key for the Google Gemini API.

1. Clone the Repository

git clone https://github.com/Saikumar1801/AI-Wellness-Hub.git
cd AI-Wellness-Hub

2. Set Up the Environment

Create and activate a Python virtual environment:

# For Windows
python -m venv venv
.\venv\Scripts\activate

# For MacOS/Linux
python3 -m venv venv
source venv/bin/activate

3. Install Dependencies

Install all required libraries from the requirements.txt file:

pip install -r requirements.txt

4. Add Your API Key

Create a folder named .streamlit in the main project directory. Inside it, create a file named secrets.toml and add your API key:

# .streamlit/secrets.toml
GEMINI_API_KEY = "YOUR_GEMINI_API_KEY_HERE"

5. Run the Application

The project is now ready. Launch the app with:

streamlit run Home.py

A new tab will open in your browser with the AI Wellness Hub. Enjoy!

🌟 What We Learned

This project was a powerful lesson in building AI that is not just accurate but useful and ethical. The journey from a simple predictive model to a transparent, feature-rich advisor taught us the importance of critical thinking, iterative development, and designing for real-world impact. We are incredibly proud of the final result—a tool that has the genuine potential to help create healthier workplaces.

Make sure to replace the placeholder links and image with your actual ones. This README will serve as a fantastic guide for anyone looking at your project.