An NLP-assisted and explainable rule-based platform for formative feedback from academic similarity reports

A study details the development and preliminary assessment of an NLP-assisted, explainable, rule-based platform designed to provide formative feedback from academic similarity reports. This platform aims to move beyond simple textual overlap identification to offer actionable guidance for student writing revision. Fifteen instructors evaluated the system's draft feedback for 200 undergraduate cases, assessing its perceived usefulness.

This development addresses a critical gap in educational technology by transforming static similarity reports into dynamic, actionable feedback mechanisms. It has the potential to enhance learning outcomes by providing students with clearer guidance on improving writing, while also improving the efficiency and consistency of instructor feedback processes.

Traditional academic similarity reports highlight textual overlap but lack functionality to determine plagiarism or guide students on writing revision.

Forward consideration, not a verified fact.