Interaction Effects of Teacher-Student Interaction Behaviors and Learning Styles in Primary School Intelligent Classrooms——An Empirical Analysis Based on Random Forest Algorithm and Social Network Analysis

Learning styles influence learners' interaction behaviors and content preferences, and also have a significant impact on learning outcomes. There is a close connection between teacher-student interaction behaviors and learning styles, and indepth exploration of their interaction effects is particularly important for implementing precise teaching interventions and constructing adaptive learning classrooms. Therefore, this paper took 180 teaching videos in primary school classroom as samples, applied the random forest algorithm to identify the interaction effect characteristics between teacher-student interaction behaviors and learning styles, and combined social network analysis to reveal their interaction network relationships. An empirical study was conducted on the interaction effects of teacher-student interaction behaviors and learning styles in “dialogue-based” smart classrooms in primary school. The results showed that the interaction effect values varied significantly across different “behavior-style” combinations, with the values of multiple interaction effect associated with philosophical learning style being relatively high. In such “dialogue-based” smart classrooms in primary school, teacher-student interaction behaviors and learning styles were not evenly distributed, nor were the interaction networks randomly connected, they were embedded within a three-dimensional relational network formed by learning styles, interaction behaviors, and social networks. The interaction network showed clear community differentiation, forming a network structure featuring both theoretical discussion orientation and practical transformation orientation coexisting, but with limited overall connectivity. Accordingly, this paper proposed the teaching adaptation strategies from three perspectives of questioning and feedback, practical transformation, and technical support. The research in this paper helped identify the interaction needs of students with different learning styles, provide reference for teachers to optimize classroom interaction design and deliver differentiated support, and promote the transformation of smart classroom teaching from experience-based judgment to data-driven improvement.