🧠 Cognifit: AI-Powered Mental Health Predictor

Cognifit is a user-friendly, AI-driven web application built with Streamlit to help users assess their mental health risk levels based on lifestyle, stress, and well-being factors. The tool provides personalized insights and promotes mental wellness awareness in an engaging way.

---

🚀 Features

✅ Homepage

• Introduction to Cognifit’s purpose
• Overview of key features and how it works
• Engaging Lottie animations for an appealing UI

✅ Mental Health Risk Assessment

• Interactive survey collecting user lifestyle and well-being details
• Data-driven risk prediction using a trained Logistic Regression model
• User-friendly progress bar to guide users through the form
• Immediate results with recommendations and next steps

✅ Contact Us

• Simple form for users to get in touch for feedback or queries

✅ Beautiful UI

• Responsive design with attractive animations and custom styles
• Sidebar navigation for easy access to pages

---

🗂 Project Structure

.
├── 1_app.py             # Main app entry point (Streamlit multipage)
├── 2_Home.py            # Home page with app intro and animations
├── 3_Services.py        # Mental health survey + model prediction
├── 4_Contact Us.py      # Contact Us form
├── pre_processing.ipynb # Notebook for data preprocessing & model training
├── logistic_model.pkl   # Trained logistic regression model

---

⚙️ Installation

1. Clone the Repository

git clone <git@github.com:Bhavay-sharma-21/Cognifit.git>
cd <Cognifit>

2. Install Dependencies

pip install -r requirements.txt

A typical requirements.txt might include:

streamlit
streamlit-lottie
pandas
scikit-learn
joblib
requests

---

▶️ Running the App

Run the main app:

streamlit run 1_app.py

Or, launch a specific page for testing, e.g.:

streamlit run 2_Home.py

---

🧠 How the Prediction Works

• Users provide inputs through an interactive survey:

  • Age, gender, sleep habits
  • Exercise and diet patterns
  • Stress, workload, and screen time
  • Emotional and social well-being

• Data is fed into a trained logistic regression model (logistic_model.pkl) to classify mental health risk as:

  • Low Risk
  • Moderate Risk
  • High Risk

Note: This is a predictive assessment tool and does not replace professional mental health advice.

📩 Contact

Feel free to reach out via the in-app Contact Us page for feedback or questions.