Exploration on the Reform and Practice of Civic Science Course Examination in Higher Education Institutions Based on Linear Regression Algorithm

Abstract

Exploring the examination reform of Civics in higher education institutions is to fully reflect the nurturing function of Civics. This paper introduces the linear regression algorithm, elicits the one-dimensional linear regression analysis model, and uses least squares to estimate the estimated value of the model. Then, based on the one-dimensional linear regression algorithm, the multiple linear regression algorithm is introduced. The algorithmic process of random forest is also introduced, illustrating that random forest is a classification result derived from majority voting using decision trees and bagged sampling methods. Based on the linear regression algorithm and the random forest model, a LR-RF model was constructed to analyze the examination reform and practice of the Civics course in higher education institutions, and the model was used to index the Civics course examination data of school X in two aspects, i.e., the content of the Civics course examination and the examination method. In terms of exam content, the average percentages of strongly agree, agree, disagree, and unsure are 29.53%, 26.37%, 29.28%, and 14.82%, respectively. In terms of the examination pattern, the average percentages of strongly agree, agree, disagree, and uncertain were 24.48%, 34.76%, 22.14%, and 18.62%, respectively. Based on the LR-RF model, we can effectively analyze the direction that the examination of Civics and Political Science courses in higher education institutions should be reformed, focusing on the examination content and examination mode, and focusing on reflecting the nurturing function of Civics and Political Science courses.