<![CDATA[This study addresses a gap in mathematics education research concerning the use of artificial intelligence (AI) to analyze student errors in integral calculus. Error analysis in calculus is still commonly conducted manually by lecturers, making it time-consuming, potentially subjective, and difficult to implement consistently across an entire class. Although previous studies have examined student errors in calculus, limited research has explored AI as a systematic diagnostic tool, especially among non-mathematics students. Therefore, this study aims to identify the types of errors made by students in solving integral calculus problems using an AI-based system and to examine the role of AI in supporting more accurate and targeted diagnostic assessment. The participants were 30 second-semester students from the Biology Education Study Program at Universitas Indraprasta PGRI who were taking the Integral Calculus course. This study employed a descriptive qualitative approach. An AI-based system was used to analyze students’ responses to a set of integral calculus problems and classify them into four categories: conceptual errors, procedural errors, technical errors, and errors in understanding the problem. The results showed that conceptual errors were the most dominant, occurring in 45% of students, particularly in misunderstanding the meaning of integrals and misusing integration limits. Procedural errors were found in 30% of students, technical errors in 15%, and problem-understanding errors in 10%. This study contributes empirical evidence on student error patterns, strengthens the role of AI as a systematic diagnostic tool, and provides a practical basis for lecturers to design more targeted remedial instruction in higher education settings.]]>
Copyrights © 2026
