Open access
2026
A Comprehensive Evaluation of Generative Models for Privacy-Preserving Synthetic Student Data
This work compares three synthesis paradigms, statistical, adversarial, adversarial, and diffusion-based, on two benchmarks and provides task-driven guidance for selecting a synthesizer in learning analytics.
Divine Iloh, Grace Oku, Shaozhi Jiang et al.
· IEEE Access · 0 citations