Integrating DMN and LLM for Automated Student Feedback and Support

Timely and constructive feedback is essential for supporting student learning, yet providing it at scale remains a challenge for educators. Large Language Models (LLMs) offer new opportunities to automate parts of this process, but their effectiveness depends heavily on how decision logic is formulated and maintained. When such logic is embedded directly in prompts, it can be difficult for instructors to update or adapt it over time. This thesis presents a framework that integrates Decision Model and Notation (DMN) with LLM prompting to make feedback generation more modular, transparent, and easy to refine. The approach decomposes complex evaluation rules into smaller, structured decision steps, which are then used to guide the LLM’s reasoning. The framework was applied in a graduate-level course, using student assignments and DMN models representing feedback criteria as inputs. The resulting feedback was evaluated against a Chain-of-Thought (CoT) baseline and assessed for perceived usefulness through a Technology Acceptance Model (TAM)–based survey. Results show that the DMN-guided approach improves both accuracy and consistency compared to CoT prompting, while also receiving positive responses from students. These findings suggest that combining LLMs with structured decision modeling can enhance the quality, transparency, and adaptability of automated feedback in educational settings.