Leveraging AI To Generate High-Quality Coding Practice Questions Aligned With Learning Objectives

Presented by:

Xiaojing Duan, University of Notre Dame

This poster demonstrates how to leverage AI to create high-quality coding practice questions aligned with learning objectives, reducing instructor workload while improving student learning.

Hear it from the author:

Leveraging AI To Generate High-Quality Coding Practice Questions Aligned With Learning ObjectivesXiaojing Duan, University of Notre Dame

00:00 / 01:43

Many studies have explored the potential of AI in generating educational content. The major gap is AI-generated content is often too generic. To bridge this gap, I’ve developed an AI system that can generate high-quality questions closely aligned with specific learning objectives. The system is designed using a retrieval-augmented generation approach. It allows educators to upload documents containing learning objectives. When students select a topic to practice, the system first retrieves the most relevant information from those objectives, then feeds them into the large language models and instructs them to create questions based on those specific contexts. As a result, students receive targeted practice questions as well as personalized feedback, which can potentially enhance their computational thinking and problem-solving skills. At the same time, educators can save valuable time from creating practice questions and instead focus on fostering deeper learning. Thank you for your time and interest. Please feel free to reach out if you have any questions or comments.

Key words:

AI, Automatic Content Generation, Learning Objectives Alignment

Abstract:

This poster describes the design and development of a novel AI-powered coding practice system aimed at enhancing student learning in introductory programming courses. The system leverages Large Language Models (LLMs) in combination with Retrieval-Augmented Generation (RAG) to automatically generate high-quality coding practice questions that are closely aligned with specific learning objectives. By reducing the time and effort instructors spend on creating extensive practice materials, the system allows educators to focus more on providing feedback and fostering deeper learning. For students, the availability of abundant, well-aligned practice questions supports the development of coding comprehension and computational thinking skills.

Outcomes:

1. Identify the strategies for using AI to reduce instructional workload while enhancing students’ learning outcomes.
2. Apply the best practices for leveraging AI to generate content aligned with specific learning objectives.
3. Analyze the potential benefits and challenges of integrating AI-generated content into their own teaching practices.

References:

Denny, P., Prather, J., Becker, B. A., Finnie-Ansley, J., Hellas, A., Leinonen, J., Luxton-Reilly, A., Reeves, B. N., Santos, E. A., & Sarsa, S. (2024). Computing education in the era of generative AI. Communications of the ACM, 67(2), 56–67. https://doi.org/10.1145/3624720

Hu, B., Zheng, L., Zhu, J., Ding, L., Wang, Y., & Gu, X. (2024). Teaching plan generation and evaluation with GPT-4: Unleashing the potential of LLM in instructional design. IEEE Transactions on Learning Technologies, 17, 1471–1485. https://doi.org/10.1109/TLT.2024.3384765

Kazemitabaar, M., Ye, R., Wang, X., Henley, A. Z., Denny, P., Craig, M., & Grossman, T. (2024, May). CodeAid: Evaluating a classroom deployment of an LLM-based programming assistant that balances student and educator needs. In Proceedings of the CHI Conference on Human Factors in Computing Systems (pp. 1–20). ACM. https://doi.org/10.1145/3613904.3642773

Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-t., Rocktäschel, T., et al. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33, 9459–9474.