Enhancing Academic Performance Predictions Using Deep Belief Networks: A Novel Approach for Educational Data Analysis

The capacity to predict students' performance is vital in matters related to deep learning and its relationship to educational data and, more generally, to higher education. Precise performance prediction facilitates both the selection of appropriate courses and the creation of efficient long-term study plans. It helps with predicting how well students will do in class, and it lets administration and educators keep tabs on students to support the need for training program integration. Preprocessing, feature selection, and model training are all part of the suggested method. Transforming raw data into a usable format is a crucial step in machine learning. It helps fix problems in real-world datasets and produce better outcomes. Principle Component Analysis (PCA) is frequently employed in feature selection. All through the training, DGNN was always in use. This novel approach outperforms GNN and RNN with an average accuracy of 94.27%.