Evaluation of Teaching Quality in Higher Mathematics Courses based on Artificial Neural Networks

The evaluation of teaching quality in higher mathematics courses presents significant challenges due to subjective assessment methods, varying instructional techniques, and diverse student learning capabilities. Traditional evaluation approaches often rely on student feedback and manual assessment, which are prone to bias and inconsistencies. Existing automated evaluation systems use statistical models and rule-based techniques but struggle to process complex relationships between instructional effectiveness and student performance. To overcome these limitations, this study proposes an advanced neural network-based framework that efficiently analyzes multiple teaching quality indicators, including lecture clarity, engagement levels, and student comprehension patterns. The proposed model significantly enhances evaluation precision and reduces subjective biases, leading to a more comprehensive assessment. Experimental validation on a real-world educational dataset demonstrates that the proposed method achieves an impressive accuracy of 97%, outperforming conventional evaluation models in reliability and effectiveness. By providing a robust and scalable solution, this framework contributes to the advancement of intelligent teaching evaluation systems in higher education.