<![CDATA[The low accuracy and objectivity of manual evaluation of mathematics learning in higher education impacts the quality of student research. This study aims to develop an Artificial Intelligence (AI)-based evaluation model to improve the effectiveness of assessment and the quality of mathematics education research. The method used was Research and Development (R&D) with a modified Borg & Gall model, encompassing needs analysis, design, expert validation, and limited trials. The resulting model consists of four main components: Data Input, AI Processing, Assessment Output, and Feedback System, all connected through a feedback loop. Validation results showed a "very valid" validity level (average >3.6), while trials demonstrated a 25–30% increase in assessment consistency between lecturers. The AI system was also able to detect student conceptual errors quickly and accurately. The results confirm that this model is effective in improving the objectivity, efficiency, and quality of student research. The study recommends the development of a machine learning-based feedback loop and the integration of the model into the campus Learning Management System (LMS).]]>
Copyrights © 2026
