The application of machine learning to assessment and feedback in mathematics education has received considerable attention in recent years. With the rapid advancement of artificial intelligence technologies, the integration of these innovations into educational assessments can significantly improve the accuracy and comprehensiveness of student learning assessments. This study examines how machine learning, specifically clustering algorithms such as Fuzzy C-Means (FCM), can be used to analyze student performance data to enable personalized and targeted instructional strategies. The complexity of educational research requires a multifaceted approach to evaluation, and the incorporation of fuzzy evaluation methods can address the inherent uncertainties in educational contexts. By exploring the historical evolution of educational assessment and utilizing advanced data analysis techniques, this research aims to provide insights into the development of more effective and nuanced assessment systems for mathematics education.