Optimizing Student Classification in Smart Education using Factor Analysis by Deep Boltzmann Machines and a Deep Learning Approach Improved by Bayesian optimization

Abstract

Smart education aims to personalize learning according to students' individual characteristics. As student classification is an essential task in this process, this research proposes an approach to optimize student classification using a combination of advanced techniques. Firstly, we use deep Boltzmann factor analysis to identify latent factors influencing student performance. These factors may include elements such as cognitive abilities, social characteristics, socio-economic status, and motivation levels. Secondly, we are developing an improved deep learning model optimized by the Bayesian method. The Bayesian approach allows for the incorporation of prior knowledge and management of uncertainty while optimizing the architecture and parameters of the deep learning model. The proposed approach is evaluated on a real student dataset and compared to traditional classification methods. The results show that our method outperforms existing approaches. In addition, latent factor analysis provides valuable insights into the key factors influencing student success. This research contributes to the advancement of intelligent education by providing a powerful tool for classifying students and personalizing learning.