IAFP: An Intelligent Assignment Feedback Platform for Automated Evaluation

Computer science and Engineering

The growing demand for scalable, intelligent educational tools has led to the development of the Intelligent Assignment Feedback Platform (IAFP)—a web-based solution designed to automate the evaluation of student assignments. Leveraging modern technologies such as the MERN stack, Google Cloud services, and the Gemini AI API, the platform streamlines the assessment process for both handwritten and digital submissions. Students can upload their responses in various formats including PDF, DOCX, and images. Handwritten submissions are processed using Optical Character Recognition (OCR) via the Google Vision API, enabling accurate text extraction and semantic evaluation. Once the content is digitized, the Gemini API analyzes each response for correctness, context, and completeness. The system then assigns marks and generates personalized, constructive feedback for every question. Educators benefit from a dedicated dashboard where they can create assignments, track student submissions, and monitor engagement. Real-time notifications are delivered via the SendGrid API, ensuring users remain informed throughout the process. Preliminary evaluations show that the platform delivers over 92% OCR accuracy and maintains consistency with human grading, especially for objective and semi-subjective questions. Students report improved understanding due to detailed feedback, while teachers experience a significant reduction in grading time. The IAFP not only enhances academic efficiency and transparency but also promotes student accountability and continuous learning. Designed with scalability and ease of use in mind, it offers a transformative step toward modernizing educational assessment through automation and artificial intelligence. This paper presents the architecture, key features, and real-world impact of the platform on teaching and learning practices in digital classrooms.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.