Effect of Q-matrix Misspecification on Variational Autoencoders (VAE) for Multidimensional Item Response Theory (MIRT) Models Estimation

Authors

Mahbubul Hasan, University of Memphis
Lih Y. Deng, University of Memphis
John Sabatini, University of Memphis
Dale Bowman, University of Memphis
Ching Chi Yang, University of Memphis
John Michael Hollander, University of Memphis

Abstract

Cognitive Diagnostic Model; Deep Generative Model; Educational Assessment; Interpretable Neural Network; Item Response Theory

Publication Title

Proceedings of the 15th International Conference on Educational Data Mining Edm 2022

Recommended Citation

Hasan, M., Deng, L., Sabatini, J., Bowman, D., Yang, C., & Hollander, J. (2022). Effect of Q-matrix Misspecification on Variational Autoencoders (VAE) for Multidimensional Item Response Theory (MIRT) Models Estimation. Proceedings of the 15th International Conference on Educational Data Mining Edm 2022 https://doi.org/10.5281/zenodo.6853016