Authenticity Validator for Academia: An Implementation of a Fake Degree Detection System

We have presented Authenticity Validator for Academia, a complete and modular implementation of a fake-degree detection system targeting the growing problem of academic credential fraud. By integrating OCR extraction, computer-vision-based tamper detection, and database-driven field verification in a unified FastAPI/React platform, the system achieves 93.8% accuracy, 90.5% precision, 95.0% recall, and an F1-score of 92.7% on a controlled test corpus. The average processing time of 0.45 seconds per certificate supports interactive deployment. Compared to prior work, Authenticity Validator for Academia offers a practical, open-source, institution-deployable solution that does not assume the existence of blockchain infrastructure or pre-signed digital originals. The system's modular design facilitates future enhancements, including integration of deep-learning forgery detectors (e.g., edge-attention CNNs [2]), blockchain-based certificate registries [8], mobile on-site verification, and multi-lingual OCR support. We believe Authenticity Validator for Academia constitutes a meaningful step toward scalable, trustworthy credential verification in both educational and employment contexts.

[1] Bae, J., Kim, S., & Lee, H., "AI-Based Document Forgery Detection Using Machine Learning and Image Analysis," International Journal of Artificial Intelligence Research, vol. 12, no. 2, pp. 45–58, 2025. [2] Boonkrong, S., Charoenwong, P., & Wattanapongsakorn, N., "OCR-Based Academic Certificate Verification System for Educational Institutions," International Journal of Computer Applications, vol. 185, no. 18, pp. 10–18, 2024. [3] Bhatia, R., & Mehta, A., "Blockchain-Based Academic Credential Verification Framework," IEEE Access, vol. 10, pp. 56789–56802, 2022. [4] Nife, T., Adewale, O., & Hassan, M., "Image Forensics Techniques for Document Authentication and Tamper Detection," Journal of Information Security and Applications, vol. 78, pp. 103–115, 2025. [5] Zhao, Y., Wang, X., & Chen, L., "Deep Learning-Based Document Verification and Classification Framework," Expert Systems with Applications, vol. 245, 2026.