AI PaperChecker is a Streamlit-based web application designed to automate the grading of student answer sheets by comparing them with a provided answer key. The application utilizes pdfplumber for PDF text extraction, SentenceTransformer & OpenAI's Turbo GPT 3.5 model for semantic analysis and scoring.

• PDF Upload: Easily upload the answer key and student's answer sheet in PDF format.
• Text Extraction: Extracts text from PDFs using pdfplumber.
• Question and Answer Parsing: Automatically identifies and separates questions and answers from the answer key. Question Paper should be setted in a fix format.
• Semantic Comparison: Uses a pre-trained model from sentence-transformers to compare student answers to the key.
• AI-Powered Scoring: OpenAI's GPT model assigns scores based on content similarity, length, and correctness.
• Final Report: Generates a comprehensive report with marks and percentage for each question.

To run AI PaperChecker locally, follow these steps:

1. Clone the Repository:

   git clone https://github.com/Spy-boy-07/AnswerSheetChecker.git
   cd AnswerSheetChecker
   

2. Install Dependencies::

    pip install openAI pdfplumber SentenceTransformer re streamlit
   

Replace API_KEY with your OpenAi's API key.

3. Run the Application::

   streamlit run app.py
   
   
   

1. Upload PDFs: Upload the answer key and the student's answer sheet using the file uploaders.
2. Process PDFs: Click the "Process PDFs" button to start the analysis.
3. View Results: The app will display a detailed report with scores for each answer and the overall percentage.

• Input: PDF files for the answer key and student answer sheet.
• Output: A report with individual question scores and the total percentage.

Click the image above to watch a video demonstration of the AI PaperChecker in action.