Research on the Impact of Elective Subjects of College Entrance Examination on Talent Cultivation of Higher Vocational Colleges Based on Students’ Cognitive Behavioral Characteristics

Abstract

The reform of the optional subjects of the college entrance examination under the new situation puts forward higher challenges to the talent cultivation of higher vocational colleges. In this paper, for students entering higher vocational colleges and universities after the new college entrance examination, the random forest framework is improved by mining students’ homework and quiz data with PCA dimensionality reduction. Behavioral feature extraction using Random Forest and coding definitions are proposed after feature screening. Based on general feature coding, the behavioral group differentiation extraction of higher education students was constructed, which finally provided label classification of students’ cognitive-behavioral features. After analyzing the impact of the optional subjects on the quality of higher vocational students, the learning behaviors of higher vocational students were dissected. The probability density distribution plots of students’ entry rates show that the overall probability density distribution curves of entry for different students are significantly different. The wave heights of the first, second, third and fourth academic years were 10.25, 16.48, 13.24 and 12.61, respectively, and the study behavior of senior students after the selection of subjects in the college entrance examination showed a unique pattern. In response to the behavioral performance of higher vocational students, institutions need to scientifically use assessment reports to launch talent development work.

Application of Random Forest Algorithm in English Innovative Talent Cultivation Mode in Colleges and Universities

“…Traditional teaching methods frequently struggle to satisfy students' different requirements and learning styles, necessitating a search for new alternatives. In this environment, advanced data-driven techniques, such as the Random Forest Algorithm, appear to be a promising answer [2].…”