https://github.com/tinbyte/rag_sql_quiz_generation
An AI-powered platform that dynamically generates and grades exam questions using Retrieval-Augmented Generation (RAG). It leverages NLP, document retrieval, and a user-friendly interface for seamless exam creation
https://github.com/tinbyte/rag_sql_quiz_generation
faiss javascript llama3 nextjs python python3 qcm retrival-augmented-generation sql
Last synced: 21 days ago
JSON representation
An AI-powered platform that dynamically generates and grades exam questions using Retrieval-Augmented Generation (RAG). It leverages NLP, document retrieval, and a user-friendly interface for seamless exam creation
- Host: GitHub
- URL: https://github.com/tinbyte/rag_sql_quiz_generation
- Owner: TINBYTE
- License: mit
- Created: 2025-02-10T18:55:27.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-02-10T19:05:42.000Z (about 1 year ago)
- Last Synced: 2026-01-03T17:28:22.679Z (4 months ago)
- Topics: faiss, javascript, llama3, nextjs, python, python3, qcm, retrival-augmented-generation, sql
- Language: Jupyter Notebook
- Homepage:
- Size: 12 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# **RAG System for Exam Generation**
> **An AI-powered platform to create dynamic and adaptive exams from educational content**
## **🚀 Project Overview**
The **RAG System for Exam Generation** is a state-of-the-art platform designed to simplify the process of creating and managing exams. By leveraging **Retrieval-Augmented Generation (RAG)**, the system retrieves relevant course content and uses advanced NLP techniques to generate questions dynamically. The project also includes features like automated grading and a user-friendly interface, ensuring a seamless experience for both educators and students.
---
## **📂 Features**
- **Knowledge Base Construction**:
Curated and structured database of course materials, including lecture notes, textbooks, and multimedia.
- **Document Retrieval & Preprocessing**:
Efficient retrieval mechanisms and data preprocessing to extract relevant information.
- **Dynamic Question Generation**:
AI-powered generation of multiple-choice and open-ended questions with customizable difficulty levels.
- **Automated Answer Verification & Grading**:
Instant evaluation of responses with intelligent grading algorithms.
- **Interactive User Interface**:
Built with **Next.js**, the platform offers a sleek, responsive interface for taking exams and receiving feedback.
---
## **🛠️ Tech Stack**
### **Frontend**
- **[Next.js](https://nextjs.org/)**: For creating a modern, responsive user interface.
### **Backend**
- **Python**: Core language for retrieval, question generation, and grading logic.
- **FastAPI/Flask**: For building APIs to connect the front end with backend services.
### **AI & NLP Tools**
- **Hugging Face Transformers**: For leveraging models like T5 and GPT for question generation.
- **Elasticsearch**: For efficient document retrieval.
- **Sentence-BERT**: For semantic similarity in answer grading.
### **Database**
- **PostgreSQL**: To store and manage structured data.
---
## **📚 Project Structure**
```
RAG-System-Exam-Generation/
├── frontend/ # Next.js frontend code
├── backend/ # APIs, NLP pipelines, and grading logic
├── data/ # Knowledge base data (lecture notes, textbooks, etc.)
├── models/ # Pre-trained and fine-tuned NLP models
├── scripts/ # Utility scripts for data preprocessing and testing
└── README.md # Project documentation
```
---
## **🌟 How It Works**
1. **Knowledge Base Construction**:
Upload structured course materials into the system.
2. **Document Retrieval**:
The RAG system retrieves relevant content using semantic search.
3. **Question Generation**:
AI models generate exam questions tailored to the retrieved content.
4. **Answer Verification & Grading**:
Student responses are automatically graded based on predefined algorithms.
5. **User Interaction**:
Students access exams via a user-friendly interface and receive instant feedback.
---
## **🤝 Team Members**
- **Nourdin**: Knowledge Base Construction
- **Nordin & Youssef**: Document Retrieval & Preprocessing
- **Youssef**: NLP for Question Generation
- **Abdelfattah Bouhlali**: Answer Verification, Grading, and User Interface
---
## **💻 Installation & Setup**
1. Clone the repository:
```bash
git clone https://github.com//RAG-System-Exam-Generation.git
cd RAG-System-Exam-Generation
```
2. Install backend dependencies:
```bash
pip install -r backend/requirements.txt
```
3. Run the backend server:
```bash
python backend/app.py
```
4. Start the Next.js frontend:
```bash
cd frontend
npm install
npm run dev
```
5. Access the platform at `http://localhost:3000`.
---
## **🎯 Future Plans**
- Expand question types (e.g., drag-and-drop, matching).
- Add support for adaptive exams based on student performance.
- Implement multilingual support for diverse educational contexts.
---
## **📜 License**
This project is licensed under the MIT License. See the `LICENSE` file for details.