VarFA: A Variational Factor Analysis Framework For Efficient Bayesian Learning Analytics
This work addresses the need for efficient uncertainty estimation in adaptive testing scenarios for educational data mining, though it is incremental as it extends existing factor analysis models with variational inference.
The authors tackled the problem of computationally expensive Bayesian inference for factor analysis in educational data mining by proposing VarFA, a variational inference framework that efficiently outputs uncertainty estimates, demonstrating its efficacy on synthetic and real datasets with scalability to large datasets.
We propose VarFA, a variational inference factor analysis framework that extends existing factor analysis models for educational data mining to efficiently output uncertainty estimation in the model's estimated factors. Such uncertainty information is useful, for example, for an adaptive testing scenario, where additional tests can be administered if the model is not quite certain about a students' skill level estimation. Traditional Bayesian inference methods that produce such uncertainty information are computationally expensive and do not scale to large data sets. VarFA utilizes variational inference which makes it possible to efficiently perform Bayesian inference even on very large data sets. We use the sparse factor analysis model as a case study and demonstrate the efficacy of VarFA on both synthetic and real data sets. VarFA is also very general and can be applied to a wide array of factor analysis models.