A Study on Optimizing Teaching Assessment Methods in Mathematical Modeling Courses by Combining Big Data Analytics Techniques

In this paper, starting from the teaching practice of mathematical modeling courses in colleges and universities, the application of fuzzy mathematical theory in teaching evaluation systems is studied, and the fuzzy evaluation model established by using fuzzy mathematics is analyzed and designed in depth, and the fuzzy matrix arithmetic process is given to realize the construction of a fuzzy comprehensive judging model. Given the substantial volume of data gathered to assess the teaching performance of teachers in mathematical modeling courses, direct classification is a challenging task. This paper introduces an unsupervised K-means algorithm that quickly classifies all the collected data and builds a teaching evaluation system for university mathematical modeling courses using relevant web development techniques. This paper's teaching evaluation system is helpful for teachers to improve their teaching situation and adjust their teaching strategies in a timely manner. The cluster analysis organized the teachers into four categories: excellent, good, moderate, and poor. The evaluation results were obtained after evaluating the fuzzy comprehensive evaluation model. The system's evaluation results in this paper demonstrate a confidence level of more than 0.85. It shows that the system is more reasonable, and its use accelerates the efficiency of teaching evaluation in the mathematical modeling course in colleges and universities, which greatly improves teaching management.