StressSenseAI – Mental Stress Detection System

StressSenseAI is a machine learning-powered web application designed to assess and analyze mental stress levels using a structured questionnaire and predictive modeling.

It combines psychological assessment with ML-based classification, interactive visualizations, and report generation to provide meaningful insights into user stress patterns.

---

🧠 Key Features

• 18-question stress assessment (psychological, behavioral, cognitive)

• Machine Learning-based stress prediction using trained models

• Voice-enabled interaction (speech-to-text & text-to-speech)

• Interactive visualizations:

  • Gauge chart (stress level)
  • Radar chart (category-wise analysis)
  • Bar chart (response breakdown)

• Personalized stress management recommendations

• Role-based insights via dashboard

• SQLite-based response tracking

• Downloadable PDF reports

---

⚙️ Tech Stack

Backend:

• Python (Flask-based web app)

Frontend:

• HTML (rendered via Flask)
• Browser APIs for speech recognition & TTS

Machine Learning:

• Scikit-learn
• Joblib (model serialization)

Database:

• SQLite

Visualization & Reports:

• Matplotlib
• ReportLab

---

🤖 Machine Learning Details

• Model Type: (e.g., Logistic Regression / Random Forest — update this)
• Input: Questionnaire responses
• Output: Stress Level Classification (Low / Medium / High)
• Model stored in: ml_model/

---

📁 Project Structure

.
├── stress_app.py        # Main application server
├── ml_model/            # ML models and training utilities
├── data/                # Runtime data files
└── stress_data.db       # SQLite database (generated at runtime)

---

🚀 Getting Started

1. Create Virtual Environment

python -m venv .venv

2. Activate Environment

Windows:

.\.venv\Scripts\Activate.ps1

3. Install Dependencies

pip install matplotlib numpy pandas reportlab scikit-learn joblib

---

▶️ Run the Application

python stress_app.py

Open in browser:

http://localhost:5050

Dashboard:

http://localhost:5050/dashboard

---

📊 Output & Reports

• Real-time stress level prediction
• Visual insights using charts
• Downloadable PDF report for each assessment
• Historical tracking via dashboard

---

⚠️ Important Notes

• stress_data.db is runtime-generated → do NOT commit

• Use .gitignore for:

  • .venv/
  • __pycache__/

• Large ML models → use Git LFS or provide download instructions

• This system is not a medical diagnostic tool

---

🎯 Project Goal

To provide an accessible and intelligent system for early stress detection using machine learning and user-friendly interaction.