This project aims to streamline the admissions process at Ashesi University by reducing the workload on admissions recruiters. Specifically, the project implements an optical character recognition (OCR) model to automate the extraction and analysis of student grades from the West African Senior School Certificate Examination (WASSCE). The system leverages Google Document AI API for text and data extraction from scanned documents, and a Streamlit application serves as the user interface for admissions officers.
https://github.com/selomcaleb/Capstone-Project--40092024.git
- Google Document AI API: Extracts text and data from scanned WASSCE result sheets using advanced machine learning models, converting them into structured formats.
- Streamlit Application: Provides a user-friendly interface for admissions officers to upload result files, view extracted data, and access detailed information quickly.
- File Upload: Admissions officers upload scanned WASSCE results through the Streamlit interface.
- Data Extraction: The uploaded documents are processed by the Document AI API, extracting relevant data such as the student's name, subjects, and grades.
- Data Display: The extracted information is displayed on the application, allowing admissions officers to review and use the data efficiently.
- Python 3.x
- Streamlit
- Google Cloud SDK (for Document AI)
- python3 capstone_ocr.py
- streamlit run app.py