Design and Evaluation of an AI-Supported Structured Project-Based Learning Model in an Undergraduate Deep Learning Technology Course
*Author to whom correspondence should be addressed.
Generative artificial intelligence (AI) can support project-based learning (PBL), but without instructional structure, it may facilitate task completion without sustained analytical engagement. This study developed and implemented an AI-supported structured project-based learning (AI-SPBL) model in an undergraduate Deep Learning Technology course. The model treated AI as part of a structured project process and required a project proposal at the midterm assessment to make students’ reasoning visible before implementation. A historical cross-cohort comparison examined 61 software engineering students across 18 group projects. Compared with the 2025 comparison cohort, which used conventional PBL, the 2026 intervention cohort produced technical reports showing broader analytical practices, while project-level scores were modestly higher and less dispersed. These findings suggest that embedding AI within a structured project process may shift its role from a task-efficiency tool toward a scaffold for more visible, evidence-based project reasoning.
[1] Yan Y, Ding R, Zhang H, et al., 2026, A Multi-Role AI Agent for Project-Driven Teaching Practice in Software Engineering. Journal of Contemporary Educational Research, 10(7): 219–224.
[2] Pu X, Zhang J, 2026, AI-Supported Project-Based Learning Reform for the Data Structures Course under the “Less Teaching, More Learning” Philosophy. Journal of Contemporary Educational Research, 10(7): 161–166.
[3] Chen Y, 2026, Research on Teaching Reform of “Introduction to Civil Engineering” Through Bidirectional Integration of Generative AI and Problem-Based Learning. Journal of Contemporary Educational Research, 10(2): 90–95.
[4] Kestin G, Miller K, Klales A, et al., 2025, AI Tutoring Outperforms In-Class Active Learning: An RCT Introducing a Novel Research-Based Design in an Authentic Educational Setting. Scientific Reports, 15: 17458.
[5] Nikolic S, Quince Z, Lindqvist AL, et al., 2025, Project-Work Artificial Intelligence Integration Framework (PAIIF): Developing a CDIO-Based Framework for Educational Integration. STEM Education, 5(2): 310–332.
