Dynamic Question Generator (B.Tech Final Year Project)

A Flask-based system for Teachers and Students to upload questions (CSV/PDF), generate dynamic MCQ tests, assign them to students, conduct assessments, and view analytics with charts. SQLite is used as the database. The UI features a modern Liquid Glass style with Dark/Light mode.

Features

• Teacher and Student roles with authentication
• Upload questions from CSV or PDF (auto BL level/topic/difficulty tagging heuristics)
• Create tests with filters; large pool per test, per-student randomized question sets
• Assign tests to students
• Students take tests; automatic grading and results
• Analytics on Teacher and Student dashboards with charts
• Liquid Glassy UI, responsive, Dark/Light mode

Tech Stack

• Backend: Flask, Flask-SQLAlchemy
• DB: SQLite
• NLP/Parsing: pandas, pdfplumber (basic heuristics)
• Frontend: HTML/CSS/JS, Chart.js

Project Structure

D:\\DQG_Project
├─ app.py
├─ database.py
├─ nlp_processor.py
├─ requirements.txt
├─ README.md
├─ instance\ (created automatically)
├─ uploads\
├─ static\
│  ├─ css\style.css
│  └─ js\main.js
└─ templates\
   ├─ base.html
   ├─ login.html
   ├─ register.html
   ├─ teacher\
   │  ├─ dashboard.html
   │  ├─ upload_questions.html
   │  ├─ create_test.html
   │  ├─ assign_test.html
   │  └─ view_results.html
   └─ student\
      ├─ dashboard.html
      ├─ take_test.html
      └─ results.html

Setup

1. Python 3.10+
2. Create venv and install deps:

python -m venv .venv
. .venv/Scripts/activate
pip install -r requirements.txt

3. Run the app:

python app.py

4. Open http://localhost:5000

Default teacher account: admin / admin123

Uploading Questions

• CSV format headers: Question, Option A, Option B, Option C, Option D, Answer, BL Level (optional: Topic, Difficulty)
• You can use your provided DSA_Questions.csv and Machine_Learning_Questions.csv directly.
• PDF: basic heuristic extracts question-like lines and generates simple distractors.

Notes

• Tests create a pool of filtered questions. Each student gets a randomized subset of size num_questions from that pool, ensuring different sets across students.
• Chart.js is used via CDN.

Security

• Passwords hashed with Werkzeug.
• Flask session-based auth; for production, use environment secrets and HTTPS.

License

MIT