Back to Portfolio

VisioMark - AI Powered MCQ Sheet Grading System

COMPLETION TIME

32 WEEKS

FEE CHARGED

₵0

BUILT WITH

React, Node.js, TensorFlow.js, OpenCV, Tauri

CLIENT

University Supervisor

VisioMark - AI Powered MCQ Sheet Grading System
Scroll to see more

Project Overview

VisioMark is a sophisticated desktop application that automates the grading process for multiple-choice question (MCQ) examination sheets. Using computer vision and machine learning algorithms, the system can process, and grade hundreds of answer sheets in minutes, providing detailed analytics and insights for educators

The Challenge

The manual grading of MCQ examination papers is time-consuming, error-prone, and resource-intensive. The National Education Board needed a scalable solution that could accurately process thousands of examination sheets while providing comprehensive analytics and maintaining data security.

The Solution

I design & developed a full-stack desktop application with an intuitive dark-themed interface that guides users through the entire grading workflow. The system uses advanced computer vision algorithms to detect and analyze answer sheets, machine learning for answer recognition, and provides real-time feedback with visual indicators for correct and incorrect answers. The dashboard includes comprehensive analytics, performance metrics, and exportable reports

The Results

VisioMark has processed over 50,000 examination sheets since released, reducing grading time by 95% and achieving a 97.8% accuracy rate compared to manual grading. The system has saved approximately 2,000 teacher hours per examination cycle and provided valuable insights that have helped improve question quality and exam design.

Client Testimonial

"VisioMark has transformed our examination process. What used to take weeks now takes hours, with even greater accuracy. The analytics provided have been invaluable for understanding student performance patterns and improving our assessment methodologies. The intuitive interface made adoption seamless across our diverse user base."

Prof. Emmanuel Akowuah, University Supervisor

Project Details

CLIENT

University Supervisor

YEAR

2024

ROLE

Full-Stack Developer & AI Implementation Specialist

COMPLETION TIME

32 WEEKS

FEE CHARGED

₵0