Exploring 2D Geometric Shape Classification Using AI-Driven Feature Tables in Mathematics

Abstract

This study explored the effectiveness of an AI-integrated instructional task designed to enhance preservice teachers' understanding of the features and hierarchical relationships of 2D geometric shapes. Originally developed and tested in online K-12 professional development settings, this intervention was adapted for in-person preservice teacher education context in this study. Data were collected from 17 preservice teachers through demographic surveys, pre- and posttests using the Van Hiele geometry framework, hierarchical diagram tasks, feature table creation during the intervention, and postintervention reflections. Findings indicated a statistically significant improvement in the accuracy and complexity of postintervention hierarchical diagrams, along with a descriptively higher mean score on the posttest of Van Hiele geometry content knowledge. Postintervention diagram classifications revealed a greater number of participants achieving the highest level of understanding of hierarchical relationships. Thematic analysis of participants' reflections suggested an increased awareness of AI integration in teaching and a deeper conceptual understanding of classification and hierarchy. This study highlights a practical approach for future educators to incorporate AI concepts into mathematics instruction, supporting the connection between abstract geometric ideas and real-world applications.