Enhancing Institutional Quality Assessment in Higher Education Using LSTM-NMPSO Hybrid Model

Advancements in educational assessment methodologies, driven by high-speed data networks, have enabled the efficient management and analysis of large datasets, replacing traditional testing methods. Even though they are frequently used, traditional statistical methods have the potential to incorporate biases into assessments of the caliber of universities. To address these limitations, the application of automated technologies is necessary for identifying key factors influencing institutional quality. Developing effective educational programmes in higher education requires quality assurance. Academic performance evaluation using Machine Learning (ML) and Artificial Intelligence (AI) techniques yields more accurate predictive models than traditional methods. This research proposes a hybrid approach that integrates Long Short-Term Memory (LSTM) neural networks with Novel Modified Particle Swarm Optimization (NMPSO) to optimize model architecture, enabling more precise and unbiased assessments of institutional quality. The objective of the proposed methodology is to improve the objectivity and reliability of institutional quality evaluations in higher education.