RLBSA-based Academic Information System Optimization for Student Performance Prediction

Abstract

Academic information systems play an important role in student data management and data-driven decision-making. However, traditional analysis methods such as Decision Tree (DT) and Support Vector Machine (SVM) often suffer from limitations in prediction accuracy and processing efficiency. This research develops an Academic Information System based on Random Leapfrog Band Selection Algorithm (RLBSA) to improve student performance prediction accuracy and academic data processing efficiency. The system adopts Google Firestore (NoSQL) architecture based on cloud computing, which enables large-scale data management with low latency and high scalability. Experimental results show that the RLBSA-based model achieves a prediction accuracy of 94.3%, higher than that of SVM (89.7%) and DT (87.4%). In terms of efficiency, the RLBSA-based system reduces data processing time by 40% compared to traditional methods, making it faster in handling large-scale academic datasets. In addition, scalability testing shows that the system is capable of handling up to 1,500 simultaneous users with an average latency below 250 milliseconds, proving its superiority in cloud-based academic environments. This research contributes to the development of data-driven academic evaluation systems, algorithm optimization in student performance analysis, as well as the application of cloud technology in academic information systems. The implications of this research open up opportunities for further integration with deep learning and reinforcement learning to improve accuracy and efficiency in academic decision making.