ASTRA

An AI Model for Understanding Math Strategies

Our research lab at the University of Memphis in collaboration with Carnegie Learning is developing AI technologies to better understand Math learning. We are currently analyzing how students apply math strategies within MATHia (formerly Cognitive Tutor), an Intelligent Tutoring System that is part of a core, blended curriculum used widely in U.S. schools.

Papers

Abisha Thapa Magar, Stephen E. Fancsali, Vasile Rus, April Murphy, Steve Ritter, Deepak Venugopal. Analyzing Strategies in MATHia with BERT, Artificial Intelligence in Education (AIED) 2025. [pdf]
Abisha Thapa Magar, Stephen E. Fancsali, Vasile Rus, April Murphy, Steve Ritter, Deepak Venugopal. "Can A Language Model Represent Math Strategies?": Learning Math Strategies from Big Data using BERT, International Learning Analytics and Knowledge Conference (LAK), 2025. [pdf]
Abisha Thapa Magar, Stephen E. Fancsali, Vasile Rus, April Murphy, Steve Ritter, Deepak Venugopal. Learning Representations for Math Strategies using BERT, published at ACM Learning @ Scale 2024. [pdf]

Presentations

Poster at AERA 2025
- https://aera25-aera.ipostersessions.com/default.aspx?s=19-AD-E5-B0-78-EB-BC-70-51-B2-77-D5-9E-76-70-D0&guestview=true
Using AI to Understand the Impact of Math Strategy Use, AIMS Winter Meeting (Steve Ritter, Carnegie Learning), Deepak Venugopal (U of M) 2024. [pdf]
Learning to Represent Math Strategies, UM Cognitive Science Seminar, 2024.

Software

AI Models

Path Analysis Tool

Data

Links available soon!

Page updated

Google Sites

Report abuse