Exploring the patterns in the undergraduate curriculum from the perspective of networks

As time goes on, mountains of files recording undergraduate students’ academic performance are left silently in the archive rooms of universities and colleges. How to apply these education outcome data to evaluate the rationality and effectiveness of the cultivation program and to provide appropriate advices for curriculum design and redesign is of crucial practical value in higher education. Here we reveal patterns of the undergraduate curriculum mining from students’ academic score data from networks perspectives. Firstly, by decreasing the threshold of distance between courses, the peripheral-core structure is filtered quantitatively. Secondly, with the auxiliary of minimum spanning tree and considering the semester that courses are given, we can judge the rationality of the curriculum scheduling and proper advices followed naturally. Finally, the hidden gender difference are uncovered visually and quantitatively. With the guide of our findings, necessary quantitative monitor, feedback and adjustment can be suggested during the reform and innovation of higher education mining from students’ education outcome data. The method we presented appears to be a useful tool to explore educational data with possible future implications on education and other fields.