Skip to main content Skip to navigation

Improving mathematical questioning in teacher training

Project Overview

The document explores the integration of generative AI in education, specifically focusing on a text-based interactive conversational agent aimed at enhancing teachers' mathematical questioning skills through AI-driven simulated classroom environments. By employing human-centered design, deep learning, and uncertainty quantification, this system seeks to provide effective teacher training while recognizing inherent limitations such as data scarcity and challenges related to user expectations. The findings underscore the significance of human-AI collaboration in educational settings, highlighting that while AI can offer innovative tools for educators, successful implementation requires careful consideration of these challenges. Overall, the document illustrates the potential of generative AI to transform educational practices by fostering improved pedagogical skills and adapting to teachers' needs in a supportive manner.

Key Applications

Text-based interactive conversational agent for practicing mathematical questioning

Context: Teacher training for effective mathematical questioning strategies

Implementation: Developed a dialogue system that allows teachers to practice questioning techniques with a simulated student using AI

Outcomes: High user satisfaction and improved conversation success rate; teachers felt they were able to practice and enhance their questioning skills effectively

Challenges: Limitations of conversational agents in modeling complex dialogues, data scarcity, and user expectation misalignment

Implementation Barriers

Model and Data Limitations

Deep learning models used in conversational agents can fail due to reliance on spurious relationships in datasets, leading to challenges in educational contexts where data is limited.

Proposed Solutions: Incorporate human-AI collaboration to mitigate data scarcity and model fragility.

User Expectation Limitations

Difficulty in setting accurate user expectations for task-specific conversational agents which can lead to a mismatch in perceived capabilities.

Proposed Solutions: Improve user feedback mechanisms and clarify the limitations and capabilities of the conversational agents.

Project Team

Debajyoti Datta

Researcher

Maria Phillips

Researcher

James P Bywater

Researcher

Jennifer Chiu

Researcher

Ginger S. Watson

Researcher

Laura E. Barnes

Researcher

Donald E Brown

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Debajyoti Datta, Maria Phillips, James P Bywater, Jennifer Chiu, Ginger S. Watson, Laura E. Barnes, Donald E Brown

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies