Generative AI Project: MCQ Generator using OpenAI, Langchain Streamlit

Author: Muhammad Adil Naeem

Overview

The MCQ Generator is a comprehensive project designed to generate multiple-choice questions (MCQs) from provided text, evaluate the complexity of these questions, and present them through a user-friendly web interface. This project leverages LangChain and Streamlit for its core functionality.

Generative.Ai.Project.-.Visual.Studio.Code.mp4

Table of Contents

• Generative AI Project: MCQ Generator using OpenAI, Langchain Streamlit
  • Author: Muhammad Adil Naeem
  • Overview
  • Table of Contents
  • Project Structure
  • Features
  • Installation
  • Usage
  • Configuration
  • Data
  • Experiments
  • Acknowledgements

Project Structure

MCQ-Generator-main/
├── .gitignore
├── README.md
├── doubt.txt
├── mcq_training_data.txt
├── requirements.txt
├── response.json
├── setup.py
├── streamlit.py
├── experiments/
│   ├── machine_learning_quiz.csv
│   └── mcq.ipynb
└── src/
    ├── __init__.py
    └── mcqgenerater/
        ├── MCQgenerater.py
        ├── __init__.py
        ├── logger.py
        └── utils.py

Features

• MCQ Generation: Generate MCQs from provided text using advanced natural language processing techniques.
• Complexity Evaluation: Assess the complexity of the generated MCQs.
• Web Interface: User-friendly web interface to interact with the MCQ generator.

Installation

To install the necessary dependencies, run the following command:

pip install -r requirements.txt

Usage

To use the MCQ generator, run the streamlit.py script:

streamlit run streamlit.py

Configuration

You can configure various aspects of the project in the setup.py file and adjust logging settings in src/mcqgenerater/logger.py.

Data

The project includes example training data (mcq_training_data.txt) and a sample response file (response.json).

Experiments

The experiments directory contains a Jupyter notebook (mcq.ipynb) and a CSV file with machine learning quiz data (machine_learning_quiz.csv).

Acknowledgements

We would like to thank the developers of LangChain and Streamlit for their excellent tools and frameworks. Also ineuron, for giving us the opportunity to work on this project.