Toward Developing Responsible Conversational AI With Induction, Concretization, and Exemplification Strategies for Math Learning

• In Event: Thursday Roundtable Session 12:40 pm
  In Roundtable Session: AI and Computational Thinking in Action (Table 4)

Abstract

Algebra is an essential yet challenging subject. A promising approach to automatically support students' Algebra learning at a large scale is conversational artificial intelligence (ConvAI). While prior educational studies on ConvAI have invited the adoption of it in education, responsibility issues and a lack of empirical understanding have emerged from the literature. This study aimed to fill the gap by designing and developing a ConvAI using responsible generation techniques to support students' online Algebra learning. The experimental results showed that students demonstrated a high level of technology acceptance toward the responsible ConvAI. Analyzed with their eye-tracking data, students in the experimental group showed a different attention trajectory from those in the control group, indicating the engagingness of the ConvAI.

Authors

• Chenglu Li, University of Utah

• Wanli Xing, University of Florida