Ryan Baker on Using Educational Data Mining to Detect the Moment of Student Learning

| March 23, 2011

Ryan Baker from Worcester Polytechnic will join us to discuss his work on educational data mining.

Increasingly, students’ educational experiences occur in the context of educational technology, creating opportunities to log student behavior in a fashion that is both longitudinal and very fine-grained. These data are now available to the broad learning sciences research community through large public data repositories such as the Pittsburgh Science of Learning Center DataShop (cf. Koedinger et al, 2008). In this talk, Professor Baker will discuss how the emerging Educational Data Mining community is combining these data sources with data mining methods in order to scalably use this data to make basic discoveries about learners and learning.

Professor Baker will illustrate the use of Educational Data Mining methods through two recent analyses his group has conducted. The first investigates how small-scale details in educational software design have large impacts on how seriously
students take the software, and in turn how much they learn, using data from one school of students using educational software for Algebra for an entire year. The second analysis uses data from students using educational software for Middle School Mathematics to develop models that can predict not only whether a student has learned, but also when the learning occurs. This model in turn allows researchers to develop models of the “spikiness” of student learning that can be used to investigate the differences between skills learned through insight and “eureka” moments, and skills learned more gradually.

Ryan S. J. d. Baker is Assistant Professor of Psychology and the Learning Sciences
at Worcester Polytechnic Institute, with a collaborative appointment in Computer
Science. He graduated from Carnegie Mellon University in 2005, with a Ph.D. in
Human-Computer Interaction. He is the former Technical Director of the Pittsburgh Science of Learning Center DataShop, the world’s largest public
repository for data on the interaction between students and educational software. He is Associate Editor of the Journal of Educational Data Mining, and in summer 2010, he was the conference chair of the Third International Conference on Educational Data Mining. He was also co-editor of the Handbook of Educational Data Mining, published in the Fall of 2010. His article ‘Detecting Student Misuse of Intelligent Tutoring Systems’ was, as of September 2009, the third-most cited article in the first ten years of Educational Data Mining. He received the Best Paper Award at the Intelligent Tutoring Systems Conference in 2006, and received the Best Oral Presentation Award at the Intelligent Tutoring Systems Conference in 2010.