Analyzing the Effect of Blended Instructional Models for Accounting on Academic Achievement Based on Multilayer Perceptron Modeling

Abstract

Blended teaching provides a flexible and personalized solution for teachers' teaching by integrating online resources and offline interactions. The purpose of this paper is to explore the impact of accounting blended teaching mode on academic achievement, for this purpose, data mining of students' online learning behaviors is carried out, and then the Gray Wolf Optimization Algorithm is improved to propose a multilayer perceptron model based on IGWO, and the performance of the model is examined, and it is applied to the practice of academic achievement prediction. The experimental results show that the proposed method can obtain higher classification accuracy, and the mean classification accuracy AVG is about 1.04%, 1.08% and 1.28% higher than that of the GWO algorithm, FF algorithm and FPA algorithm, respectively, and the standard deviation of the indexes is small, which is of good stability. Introducing the dynamically changing learning style identification results into the blended teaching mode academic achievement prediction model, its prediction accuracy is improved from 0.819 to 0.847, which is a significant increase in accuracy. The prediction model constructed in this paper has important practical reference value for improving the accounting blended teaching model.