Artificial Intelligence for Evaluation in Education: Enhancing Handwritten Answer Sheet Assessment
International Research Journal of Modernization in Engineering Technology and Science · 2025
The evaluation of handwritten answer sheets in education is a critical but often subjective and time-consuming process.Inconsistencies in grading can arise from factors like evaluator bias and handwriting variability, impacting fairness and efficiency.This paper investigates how a multimodal Artificial Intelligence (AI) system can address these challenges.We present a prototype system that utilizes a single, unified AI model to perform both Optical Character Recognition (OCR) and semantic evaluation of handwritten text.A case study was conducted where two different handwritten answer sheets, containing identical correct answers for a computer science examination, were processed.The results show that the AI system awarded identical scores to both sheets, demonstrating a high degree of grading consistency.Furthermore, the system generated detailed, constructive feedback for each answer, highlighting its potential to not only automate but also enrich the assessment process.This study suggests that modern multimodal AI offers a promising pathway toward more equitable, consistent, and efficient educational evaluation.