作者: Kimberly Ferguson , Ivon Arroyo , Sridhar Mahadevan , Beverly Woolf , Andy Barto
DOI: 10.1007/11774303_45
关键词:
摘要: This paper describes research to analyze students' initial skill level and predict their hidden characteristics while working with an intelligent tutor. Based only on pre-test problems, a learned network was able evaluate students mastery of twelve geometry skills. model will be used online by Intelligent Tutoring System dynamically determine policy for individualizing selection problems/hints, based learning needs. Using Expectation Maximization, we the parameters several Bayesian networks that linked observed student actions inferences about unobserved features. Information Criterion different models. The contribution this work includes best network, whereas in previous work, structure fixed.