The Application of New Educational Concepts in Digital Educational Media. (21st November 2022)

• Record Type: Journal Article
• Authors: Yang, Chun
• Other Names: Zhou Tao Academic Editor.
• Is Part Of: Advances in multimedia. Volume 2022(2022)
• Publication Date: 2022-11-21
• Subjects: Multimedia systems -- Periodicals; Computer networks -- Periodicals; Multimédia; Réseaux d'ordinateurs; Computer networks; Multimedia systems; Periodicals; 006.7
• DOI: 10.1155/2022/1012803
• Languages: English
• ISSNs: 1687-5680

Abstract

Abstract : With the development and progress of society, great changes have taken place in educational concepts and teaching models compared with the past. Faced with the new educational concept of advocating diversified teaching modes and all-round talent training, the teaching space under the traditional single-fixed teaching mode is insufficient. The field of digital media education is a form of education with the development of information technology. The purpose of this paper is to find out the construction form of teaching space under the new educational concept that adapts to the development of the social era and to respond to the constantly updated and emerging educational concept and teaching mode. The process is as follows: based on the collected education data, mining the specific factors that will affect the application ability of teachers' digital education resources and building a multiple machine learning regression model using these objective and significant features to predict the score of teachers' digital education resources application ability. Through the comparison and optimization of model performance, a more suitable prediction method was found. MSE, MAE, RMSE, and MAPE are used as performance evaluation indicators to compare the performance of each model. It is found that there are multilayer linear regression < mild gradient advance < extreme gradient advance < random forest in each indicator. In addition, in the two integration models, bagging idea represented by the random forest is more suitable for this group than two gradient boosting.