Development of ideological and political education course field dimensions based on multi-data chain network

Abstract

Abstract This paper firstly classifies the ideological and political education course field dimensions, studies the specific composition of the learning field in three dimensions: information, technology and emotion, and discusses the necessity of learning field optimization in secondary school Civics class. Secondly, on the multivariate data chain network, the MIFS model is proposed to extract the ideological and political teaching feature values, followed by the LSTM neural network to train the information related to students’ Civics classroom contents and the average function of adaptive excitation function is used to replace the excitation function in the LSTM neural network to achieve the effect of improving the output accuracy. Then, the characteristics of learners and learning resources are parametrically represented, followed by an experimental analysis of ideological and political education based on multivariate data chain networks using evaluation indexes based on the data set. The results show that there is no difference between the MAE values of CNN and LSTM except for sample one, but the MSE values of LSTM are both lower than CNN, and LSTM is more stable than CNN, indicating that it is reasonable to choose LSTM when ideological and political education based on multivariate data chain network. This study provides a reference and reference for future field optimization in ideology and political science classes.