Gyanankana

Proposing a Automated Assessment Generation Platform for ALL {K-12, Higher Ed, Skilling, Hiring}

Automated Question Generation with OpenAI and Gradio

This repository contains a Python script that demonstrates how to generate questions using OpenAI's GPT-3 and provides a web-based user interface for generating questions based on a given passage.

Table of Contents

• Getting Started
  • Prerequisites
  • Setting Up OpenAI API Key
• Usage
  • Generating Questions
  • Gradio User Interface
  • Bloom's Taxonomy Levels
  • Question Types
• Contributing
• License
• Acknowledgments
• Contact

Getting Started

Prerequisites

Before running the script, you'll need to have the following prerequisites installed:

• Python 3.x
• openai library
• gradio library

You can install the required libraries using pip:

  pip install openai
  pip install gradio

Setting Up OpenAI API Key

To use OpenAI's GPT-3, you need to set your OpenAI API key. Replace "YOUR OPEN AI API KEY" in the script with your actual API key.

  api_key = "YOUR OPEN AI API KEY"

Usage

Generating Questions

The script can generate questions based on a provided passage. To generate a question, you can call the generate_blooms_question function and specify the paragraph, Bloom's Taxonomy level, and question type as parameters. The function will return the generated question.

Example:

  paragraph = "Your passage goes here."
  bloom_level = "remembering"  # Choose from "remembering," "understanding," "applying," "analyzing," "evaluating," or "creating."
  question_type = "Multiple Choice"  # Choose from "Multiple Choice" or "Multiple Select."

  generated_question = generate_blooms_question(paragraph, bloom_level, question_type)
  print(generated_question)

Gradio User Interface

The script also provides a web-based user interface using Gradio. You can interactively generate questions by running the Gradio interface. Simply run the script, and it will launch the interface.

  python automated_question_generation.py

Bloom's Taxonomy Levels

The script supports the following Bloom's Taxonomy levels for question generation:

• Remembering
• Understanding
• Applying
• Analyzing
• Evaluating
• Creating

Question Types

You can generate questions in two types:

• Multiple Choice
• Multiple Select

Contributing

• Contributions to this project are welcome. If you have ideas for enhancements, optimizations, or adaptations, please feel free to contribute.
• If you are expert in pedagogy and/or assessments for K-12 or Higher Ed feel free to drop me an email provided in the Contact section.

License

• This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• This project utilizes OpenAI's GPT-3 for natural language generation.
• Gradio is used for creating the interactive user interface.

Contact

• For any questions, suggestions, or collaborations, you can reach me at shivranjan1689@gmail.com.