Node.js MCQ App with Gemini AI

This is a Node.js backend application for generating MCQs (Multiple Choice Questions), summarizing text, evaluating answers, and more using Gemini AI. The backend is built using Express.js and leverages GoogleGenerativeAI for AI-driven content generation and evaluation.

Live: https://gen-mcq.onrender.com/

Features

• Get Output from User Prompt: Generate content based on a user-provided prompt.
• Generate MCQs: Create multiple-choice questions for given text, difficulty level, and optional image.
• Summarize Text: Provide a summary of given text with optional image association.
• Evaluate Answer: Evaluate a given answer, provide marks based on the expected answer length, and offer detailed feedback and corrections.

Requirements

• Node.js
• NPM (Node Package Manager)
• GoogleGenerativeAI API Key
• Python (for testing purposes)
• Requests library for Python (install using pip install requests)

Installation

1. Clone the repository:

   git clone https://github.com/UltronTheAI/GEN-MCQ.git
   cd GEN-MCQ

2. Install dependencies:

   npm install express dotenv @google/generative-ai fs

3. Create a .env file in the project root to store your API key:

   GEMINI_API_KEY=your_gemini_api_key_here
   

Usage

1. Start the server:

   node app.js

2. Use the following endpoints to interact with the backend:

Endpoints

1. Get Output from User Prompt

• Endpoint: /get-output
• Method: GET
• Parameters:
  • prompt (required): The user prompt to generate content.
• Response:

  {
    "output": "string"
  }

2. Generate MCQs

• Endpoint: /generate-mcq
• Method: GET
• Parameters:
  • text (required): The text for which MCQs need to be generated.
  • number_of_questions (required): The number of questions to be generated.
  • level (required, 1 to 10): The difficulty level of the questions.
  • image (optional): An image associated with the text, if applicable.
• Response:

  [
    {
      "question": "string",
      "options": ["list of strings"],
      "answer": "integer",
      "explanation": "string"
    }
  ]

3. Summarize Text

• Endpoint: /summarize
• Method: GET
• Parameters:
  • text (required): The text to be summarized.
  • image (optional): An image associated with the text, if applicable.
• Response:

  {
    "summary": "string"
  }

4. Evaluate Answer

• Endpoint: /evaluate-answer
• Method: GET
• Parameters:
  • question (required): The question text.
  • answer (required): The user's answer.
  • max_marks (required): The maximum marks for the question.
• Response:

  {
    "marks_scored": "integer",
    "explanation": "string",
    "correct_answer": "string"
  }

Python CLI for Testing

Use the provided Python script (test_backend.py) to test the endpoints.

Installation

1. Ensure Python and pip are installed.
2. Install the requests library:

   pip install requests

Usage

1. Run the script to test different endpoints:

• Get Output from User Prompt:

  python test_backend.py get-output "Explain how AI works"

• Generate MCQs:

  python test_backend.py generate-mcq "Python Basics" 5 3 --image "path_to_image.jpg"

• Summarize Text:

  python test_backend.py summarize "Photosynthesis is the process used by plants to convert light energy into chemical energy." --image "path_to_image.jpg"

• Evaluate Answer:

  python test_backend.py evaluate-answer "What is the capital of France?" "Paris" 1

Contributing

Feel free to fork the repository and submit pull requests. For major changes, please open an issue to discuss what you would like to change.

License

This project is licensed under the MIT License.