Predictive Value Chain Analysis to Improve Generalization Accuracy In Quality Assurance

Adaptive predictive analysis model utilizing Spiking Neural Networks (SNNs) to enhance quality assurance processes in Smart Campus environments. Traditional predictive models often struggle with the dynamic and complex nature of educational data, leading to inaccurate predictions and inefficient decision-making. SNNs, inspired by biological neurons, are particularly suited for processing temporal data and adapting to evolving patterns. The proposed model integrates spike-timing-dependent plasticity (STDP) and dynamic weight adjustment to continuously learn and refine predictions in response to real-time data, ensuring improved accuracy and robustness. The study uses historical data from the Internal Quality Assurance System (SPMI) over a three-year period, focusing on key performance indicators such as student enrollment trends, faculty productivity, and resource utilization. Experimental results demonstrate that the SNN-based model outperforms traditional neural networks, achieving a 20% reduction in Mean Squared Error (MSE) and a higher R-squared value. These findings indicate that the model not only enhances predictive accuracy but also offers significant improvements in operational efficiency and resource management. By providing timely and precise insights, this model supports proactive decision-making in educational institutions, enabling administrators to optimize resource allocation, monitor academic quality, and anticipate future challenges. This research contributes to the growing field of artificial intelligence in education, offering a scalable, adaptive approach to managing complex data and improving institutional performance in real-time.