Authors : Vishu; Vaibhav Rana; Varun Verma

Volume/Issue : Volume 11 - 2026, Issue 5 - May

Google Scholar : https://tinyurl.com/2ncbjdba

Scribd : https://tinyurl.com/2nbpkhce

DOI : https://doi.org/10.38124/ijisrt/26May076

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : The AI-Based Evaluation Tool for Academics is a smart platform created to improve grading efficiency and protect academic integrity in the age of artificial intelligence. As AI-generated assignments become more common, teachers struggle to identify whether student work reflects genuine effort. This project introduces a middle-layer system that automates the evaluation of multiple submission formats, including PDFs, images, and handwritten scans. The system follows four main stages: upload, extraction, de-tection, and grading. Optical Character Recognition technologies such as Google Vision API or Tesseract convert scanned docu-ments into readable text. The extracted content is then examined using fine-tuned BERT or RoBERTa models to detect patterns typical of machine-generated writing. Submissions flagged as AI-produced receive a penalty to maintain fairness. For verified human work, the tool applies keyword-based semantic similarity scoring aligned with instructor-defined cri-teria. Developed with a React frontend and Python backend, the platform streamlines assessment, reduces workload, and delivers timely, consistent feedback.

Keywords : Artificial Intelligence, Automated Evaluation, Machine Learning, Natural Language Processing, Educational Technology.

References :

1. S. Fariello, G. Fenza, F. Forte, and M. Marotta, “Distinguishing human from machine: A review of advances and challenges in AI-generated text detection,” International Journal of Interactive Multimedia and Artificial Intelligence, vol. 12, no. 2, pp. 1–25, 2025.
2. R. Gao et al., “Automatic assessment of text-based responses in post-secondary education,” Computers & Education: Artificial Intelligence, 2024.
3. I. Dada, “iAttention Transformer: An inter-sentence attention mechanism for enhanced automatic grading,” Mathematics, vol. 13, no. 18, Art. no. 2991, 2025.
4. Y. Mo, H. Qin, Y. Dong, Z. Zhu, and Z. Li, “Large language model AI text generation detection based on transformer deep learning algorithm,” arXiv:2405.06652, 2024.
5. R. Sonkar, N. Liu, D. B. Mallick, and R. G. Baraniuk, “Marking: Visual grading with highlighting errors and annotating missing bits,” arXiv:2404.14301, 2024.
6. K. Iyer, M. Ravikiran, P. Pendse, and S. Mohanty, “Towards transparent AI grading: Semantic entropy as a signal for human-AI disagreement,” arXiv:2508.04105, 2025.
7. J. Campino, “Unleashing transformers: NLP models detect AI writing in education,” Journal of Computers in Education, vol. 12, pp. 645–673, 2025.
8. A. Boutadjine, “A comparative study on the detection of AI-generated text,” ACM Transactions on Computing Education, 2025.
9. S. Aishwarya and S. Hemalatha, “Smart tracking system for academic submissions using machine learning,” in Proc. 1st Int. Conf. on AI for IoT (AI4IoT), 2023, pp. 634–639.
10. T. Kumar and R. Banerjee, “Deep learning-based OCR integration with transformer models for document understanding,” IEEE Access, vol. 11, pp. 27645–27658, 2023.
11. P. Zhang and W. Liu, “Sentence-BERT for enhanced semantic similarity in automated short-answer grading,” Educational Technology & Society, vol. 28, no. 3, pp. 55–68, 2025.
12. D. Gifu et al., “Artificial intelligence vs. human: Decoding text authen-ticity using transformer models,” Future Internet, vol. 17, no. 1, Art. no. 38, 2025.
13. “AI-generated text detection: A comprehensive review of methods, datasets, and applications,” Computer Science Review, vol. 58, Art. no. 100793, 2025.
14. P. Baker and X. Li, “AI-powered teacher assistant: Automated grading and personalized feedback with OCR and NLP,” International Journal of Engineering Research & Technology, 2026.
15. A. Ayaan, “Automated grading using natural language processing and semantic analysis,” International Journal of Education and Research, 2026.
16. “Efficient detection of AI-generated scientific abstracts with a lightweight transformer,” Journal of Computational Linguistics, 2026.