Optimizing Higher Education Curriculum Design and Teaching Content Recommendation System Using Data Mining Algorithms

Abstract

In recent years, education has been the focus of attention from all walks of life, and the education curriculum of colleges and universities is the key to cultivating students’ comprehensive quality and ability. College education needs to optimize the design of higher education courses to achieve the goal of talent training and to design a teaching content recommendation system based on students’ cognitive level. On the one hand, this study uses the support vector machine algorithm to quantitatively analyze students’ qualitative comments on the course to extract useful teaching feedback for course design optimization, and on the other hand, designs a teaching content recommendation system based on deep learning algorithm as an adjustment tool to jointly improve the teaching quality of higher education and improve the learning effect of students. The CBOW model and SVM algorithm model were used to analyze a large number of qualitative texts, and the cosine similarity of related words was calculated, such as “recommended” 0.98416, “sufficient” 0.9986 and “good” 0.99762, and the word heat map was drawn according to the word frequency. In addition, an empirical study was conducted on the teaching content recommendation system in school A. The accuracy of the recommendation results and the impact of the system on students’ academic performance were analyzed. The average accuracy of the learning resource recommendation results was 91.65%. The student’s scores were significantly improved after using the system, which proved that the design of the system was effective.