A study on algorithms for optimizing general education teaching based on student learning behavior data

Abstract

General education serves as a basic component for nurturing the comprehensive quality of students, but traditional teaching models have difficulty meeting the heterogeneity of learners and the dynamic learning needs. As learning management systems and smart classroom technologies are widely adopted, data on large-scale student learning behavior became available, which opens a new window for date-driven instructional optimization. The framework to be proposed in this study utilizes student learning behavior data through algorithm model construction. In the first step, behavior data of learning engagement, persistence, and interaction patterns are collected through multiple sources and structured to form behavioral feature representations. Then a data-driven optimization model is constructed to map learning behaviors with outcomes. It is proposed that a behavioral feedback-guided multi-objective optimization mechanism-based adaptive teaching optimization algorithm be utilized for dynamic adjustment of instructional strategies. Based on real or simulated course datasets, the results of experimental analysis prove that the improved learning performance, participation degree and instructional efficiency are evidenced in comparison with traditional teaching method by proposed theory. The findings reveal that general education frameworks can benefit from the integration of learning analytics and intelligent optimization algorithms to facilitate teachable improvement and evidence-based educational decision-making. The research informs the scalable methodological framework for intelligent teaching optimization and advances higher education pedagogic practices towards integration of educational data mining and algorithm-driven pedagogy.