Evaluation and optimization of higher education teaching quality based on big data

This paper explores the application framework of big data in the evaluation of teaching quality in higher education, focusing on data collection and integration, personalized learning path design, innovation of teaching content and methods, and support for teachers’ professional development. By analyzing students’ learning behavior data, teachers’ teaching activity data, and course resource data, this study designed and implemented a comprehensive evaluation system, including a learning ability evaluation model, a personalized recommendation system, a teaching effect prediction model, a teacher ability evaluation model, and a development indicator quantification model. Specifically, the experimental results show that each model performs well in terms of accuracy, explanatory power, and prediction efficiency: the AUC of the learning ability evaluation model reaches 0.892, the accuracy of the personalized recommendation system reaches 0.765, and the determination coefficient of the teaching effect prediction model is 0.789. These key findings not only reveal the main influencing factors of teaching quality, but also put forward suggestions for the optimization of teaching evaluation indicators, thereby providing a scientific basis for educational decision-making.