Student Behavior–Aware Intelligent Management System for English Classrooms

In recent years, integration of Artificial Intelligence (AI) into classroom management has developed as a promising approach to improve student engagement and learning outcomes, particularly in English language classrooms. Exiting research has mainly focused on content delivery, adaptive learning platforms, generalized monitoring, but most existing systems fail to capture fine-grained, individual-level student behavior and lack real-time, interpretable management capabilities. This limitation presents a major challenge for teachers aiming to maintain classroom engagement and provide timely interventions. To address these challenges, this research proposed an intelligent student behavior management framework that systematically quantifies, tracks, and interprets classroom behavior. Initially, begins with Simulation-Based Data Generation. Data Preprocessing, where multi-source behavior indicators including attention, participation, and activity features are collected, cleaned, and normalized. These features are then integrated into a Student Behavior Index (SBI), providing a unified measure of individual engagement. Students are classified into three different types of behavior states active, passive, at-risk following the construction of SBI, and temporal behavior trend analysis is performed on each of these behaviors states to measure the evolution of student engagement. Finally, an Intelligence Behavior Management Decisions Module (IBMD) incorporates the behavior states and trends of students, enabling the generation of aggregated, transparent interventions customized to each student. The proposed SB-AIMS attained better results in terms of accuracy (98.67%), precision (97.81%), recall (98.52%), F1-Score (98.54%), when compared with existing Machine Learning-Based Student Engagement Detection.