"The rigorous training provided by the ISP stimulates students' ideas for building powerful AI systems for biomedical informatics."
Ye Ye, student
Abstract: The volume and granularity of data produced by current educational technologies in K-16 presents an exciting opportunity to gain insights into student knowledge and how it is acquired. In this talk I will show how probabilistic graphical models of student learning, with roots in cognitive theory, have served as an effective platform to study learning phenomenon in digital learning environments. Probabilistic graphical models allow for a blend of machine learning and domain expertise and are well suited to capture the temporal aspects of student data. The models were used to improve the accuracy of student knowledge assessment and performance prediction by taking into account individual inferred prior knowledge and learning attributes of the student as well as attributes of the content. This approach proved effective in the 2010 KDD Cup competition on educational data mining when pitted against an ensemble of state-of-the-art machine learning approaches. I will also show how these same models have been posed to measure the pedagogical efficacy of problem orderings, individual resources in a MOOC, and remediation decisions made by the learning environment. The work presented was conducted using data from the Cognitive Tutors for Algebra and Geometry, the ASSISTments Platform, and recently, Massive OpenOnline Courses (MOOC) on the edX platform. Implications for adaptivity, affective state integration, and privacy in the big data age will be discussed.
Copyright 2009 | Web site by UMC Web Team