Speaker:

Dr. Noboru MATSUDA - North Carolina State University, United States

Moderator:

Dr. Ashwin Tudur SADASHIVA - Vanderbilt University, United States

Curatedby: Learning Analytics and Educational Data Mining (LAEDM) SIG

Date: 6 November 2024 (Wednesday)

Time: 9:00-10:00 (UTC +8)

Register before 4 November: https://us06web.zoom.us/webinar/register/WN_sO_IfOENRruFyGXgaHH-sw

Abstract

SimStudent is an interactive machine learning agent that learns cognitive skills as production rules. One of a prominent application of SimStudent is to use it as a teachable agent to study how students learn by teaching. We have been studying learning by teaching using SimStudent for middle school students to learn linear algebra. This talk will first provide an overview of our research and lessons learned from the SimStudent project. I will then introduce our latest attempt on understanding if and how teachable agent would help students learn not only how to solve equations but also conceptual knowledge to justify their reasoning. To learn algebra is to learn a complex web of algebra knowledge that contains conceptual and procedural knowledge and their relationship, which we call the connected knowledge. We hypothesize that if the teachable agent asks students (who tutor the teachable agent) to justify their reasoning *and* further provide follow up questions when students’ responses need some elaboration, then students will be engaged in deep reflective dialogue that will facilitate the acquisition of connected knowledge. To test this hypothesis, we extended SimStudent by integrating a pre-trained large language model to drive a dialogue.

Biodata

Dr. Noboru Matsuda is an Associate Professor in the Department of Computer Science at North Carolina State University.  His research interests include applications of cutting-edge artificial intelligence technologies to enhance learning, as well as to advance cognitive theories in the sciences of learning. Dr. Matsuda received an MS in Math Education from Tokyo Gakugei University (Tokyo, Japan) and a Ph.D in Intelligent Systems from the University of Pittsburgh. He has developed a number of intelligent tutoring systems primarily in STEM subjects. Dr. Matsuda has been leading the SimStudent project (www.SimStudent.org) where the research team develops an artificial intelligence that can learn problem-solving skills through guided-problem solving (aka peer tutoring). He initially started the SimStudent project when he joined CMU in 2004 for his postdoctoral training. He has since expanded the project into multiple applications including intelligent authoring, teachable agent, and learning simulation.