Trends in Artificial Intelligence Research for Geometry Education: Educational Applications and System-Oriented Approaches

Abstract

Artificial intelligence (AI) has been increasingly
introduced into mathematics education research, with
geometry emerging as a domain of particular interest due to
its strong reliance on visualization, representation, and
reasoning. Despite a growing number of studies, existing
research on AI in geometry education remains fragmented,
with limited synthesis regarding how AI applications are
positioned in relation to instructional practice. This
paper presents a focused review aimed at clarifying current
research orientations and identifying patterns in the
educational use of AI for geometry teaching and learning. A
systematic literature screening was conducted on 1,468
publications published between 2015 and 2025. Based on
predefined inclusion criteria, 80 studies were selected for
analysis. The reviewed studies were examined through
qualitative content analysis and categorized according to
their primary research orientation and educational focus.
The preliminary analysis reveals two dominant orientations.
Educationally oriented AI applications are embedded in
instructional contexts and are primarily designed to
support conceptual understanding and exploratory learning,
often under teacher guidance. In contrast, system-oriented
studies focus on the development and evaluation of AI
techniques for geometric problem solving, automated
reasoning, or representation, but rarely report classroom
implementation or pedagogical design considerations. The
findings suggest that future research should place greater
emphasis on connecting AI capabilities with well-designed
learning tasks and instructional frameworks to support
meaningful geometry teaching and learning.