Automatic Domain Model Creation and Improvement
Philip I. Pavlik Jr., Luke Eglington, Liang Zhang
Jul 02, 2021 14:10 UTC+2
—
Session PS2
—
Gather Town
Keywords: learner models, domain models, adaptive learning systems
Abstract:
We describe a data mining pipeline to convert data from educational systems into KC models. In contrast to other approaches, multiple model search methodologies (e.g., sparse factor analysis, covariance clustering) are employed and compared within a single pipeline. In this preliminary work, we describe the results of our approach on 2 datasets when using 2 model search methodologies for inferring item or KCs relations (i.e., implied transfer). The first method uses item covariances which are clustered to determine related KCs and the second method uses sparse factor analysis to derive the relationship matrix for clustering. We evaluate these methods on data from experimentally controlled practice of statistics items as well as data from the Andes physics system. We explain our plans to upgrade our pipeline to include additional methods of finding item relationships and creating domain models We discuss advantages of improving the domain model that go beyond model fit, including the fact that models with clustered item KCs result in performance predictions transferring between KCs, enabling the learning system to be more adaptive and better able to track student knowledge.