Design and Development of Artificial Intelligence-Based Smart Classroom System on B/S architecture

Combining smart technologies with learning is transforming classroom environments to make teaching more effective and learning better. By using the Browser/Server (B/S) paradigm, a powerful smart classroom system is implemented to promote real-time interaction, intelligent attendance, adaptive content provision, and learning analytics. The core objective is to deploy lightweight client interfaces through standard web browsers while delegating processing and data storage tasks to centralized servers, ensuring cross-platform compatibility and user-friendly access without dependency on specific hardware or operating systems. Current systems are plagued with poor scalability, device dependency, and disjointed learning data analysis. Conventional models are based on static content delivery and rudimentary automation without an intelligent feedback mechanism. The system proposed utilizes a Convolutional Neural Network with Bidirectional Long ShortTerm Memory (CNN-BiLSTM) for real-time facial recognition, emotion analysis, and behavior observation for personalized feedback and engagement analysis. Comparative findings show a dramatic performance improvement using the proposed CNNBiLSTM model, achieving 96.3% accuracy and a 95.8% F1score, indicating highly effective real-time student monitoring and engagement analysis in the smart classroom system based on B/S architecture. The excellence of the system is in how it combines spatial and temporal features for strong, continuous learning analytics and safe web-based accessibility.