MCQ Generator Using Langchain

Multiple Choice Question Generator using GPT-3.5 Turbo

This project is a multiple-choice question generator that leverages OpenAI's GPT-3.5 Turbo and LangChain to create customizable quizzes from input text files. Users can specify the number of questions, the subject, and the difficulty level to tailor the quiz to their needs. The app features an intuitive interface built with Streamlit, making it simple to upload a text file and configure question settings.

Features

• Text-Based Question Generation: Upload a PDF or other text file, and the app extracts content to generate questions.
• Customizable output: Users specify:
  • The number of questions
  • The subject focus
  • Difficulty level (e.g., Easy, Medium, Hard)
• Streamlit interface: A user-friendly front end that enables effortless configuration and immediate question generation. This interface can be seen in below.

Technologies Used

• OpenAI GPT-3.5 Turbo: Language model for generating high-quality multiple-choice questions.
• LangChain: Provides efficient handling of language model prompts and responses.
• Streamlit: UI framework for creating the app’s interface.
• Python: Core language used for backend logic and processing.

Installation

To run this project locally:

• 1. Clone the Repository:

git clone https://github.com/AliAfshar7/MCQ-Generator-Using-Langchain.git
cd MCQ-Generator-Using-Langchain

• 2. Set Up Environment variable: You’ll need to define an environment variable with your OpenAI API key:

export OPENAI_API_KEY="your_openai_api_key"

Also, you can define this environment variable in a .env file.

• 3. Install Requirements: Install the required dependencies using pip:

pip install -r requirements.txt

• 4. Run the Streamlit App:

streamlit run StreamlitApp.py

Usage

1. Upload a text file (such as a PDF).

2. Specify the number of questions, subject, and difficulty level.

3. Click "Create questions" to get a list of multiple-choice questions based on the given criteria.