Research on University Education Management Strategy Supported by Artificial Intelligence Big Data Technology

Abstract

Educational data mining is a practical method for unearthing the connections hidden in educational data and predicting students' academic progress. The findings of the midterm exam grades are the main data in this study's innovative machine learning-based model that forecasts undergraduate students' final exam scores. This study suggests a unique method for analysing the academic performance of distance learners utilising computer technology and machine learning approaches. Here, academic performance of distance education students is the input data that is gathered and processed for noise reduction, normalisation, and smoothing. Then, using a spatio-markov Gaussian model (SMG) and a fuzzy Q-bayes gradient vector neural network (FQBGVNN), the characteristics of the processed data were retrieved. For various student performance analysis datasets, experimental analysis is done in terms of training accuracy, average precision, recall, and MSE. In terms of how long each algorithm takes to produce the results, a comparison of the two methods is also done.