Logo Lanfrica

abdallahelahee/AI-Scan-and-Mark-Prototype

Domaine:

education

Type de record:

project
Créateur:
abd
Hôte:
AI-enhanced teacher assistant that automates CAPS marking and reporting from scanned assessments using OCR + LLMs, reducing administrative workload in South African classrooms. # Breaking the CAPS Cycle 📚🤖 ### AI Scan & Mark — Prototype This repository contains an **interactive HTML prototype** of an AI-assisted teacher assistant designed to reduce administrative workload in South African CAPS classrooms. It was developed as part of **INF3011F · Team 17 Final Project (2026)**. --- ## 🎯 Purpose This is a **prototype-only implementation** demonstrating the proposed workflow for: - Scanning learner assessments - AI-assisted marking - Generation of CAPS-compliant reports - Per-learner feedback automation It is intended for **sponsor review and system handover**, not production use. --- ## 💡 Problem Context Teachers currently spend significant time on: - Rewriting marks across multiple systems - Manual CAPS reporting - Lack of per-learner feedback generation This leads to: - High administrative burden - Reduced teaching time - Delayed feedback cycles --- ## 🧠 Solution Overview The prototype demonstrates an AI-enhanced workflow: 1. Upload/scan assessment (simulated in UI) 2. AI marking process (mocked interface) 3. Review results per learner 4. Generate CAPS class report 5. Export final report package --- ## 🖥️ Prototype Structure This project is a **single HTML-based interactive prototype**: --- ## ⚙️ Features Demonstrated - 📷 Scan/upload simulation interface - 🧠 AI marking workflow visualization - 📊 CAPS-style grading dashboard - 🔍 Per-learner feedback view - 📄 Report generation screen - 📤 Export-ready UI concept --- ## 🎓 Context Developed for: **INF3011F · Team 17 Final Project 2026** “Breaking the CAPS Cycle” Team: - Timothy Sasman - Gomolemo Modisane - Abdallah Elahee - Mateo Montoya-Pelaez - Rory Cupido Sponsor: - Charles Mayosi - Ntwanano Mabundza --- ## 🚧 Important Note This is a **front-end prototype only**: - No real AI model is connected - No backend processing exists - All outputs are simulated for demonstration purposes --- ## 🏁 Goal To communicate the **user workflow, system concept, and feasibilit …