The complex new media environment has brought greater challenges to the education and management of college students in colleges and universities, and put forward higher requirements for the work of college students. In this paper, based on the Apriori algorithm of association rules, we extracted frequent itemsets from students’ performance data sets, and mined the association rules between students’ behaviors and the risk of failing the exams according to the confidence level and support level. The potential influence variables of students’ innovation and entrepreneurship ability are explored through factor analysis, the intrinsic connection between the original variables is analyzed, and structural equations are used to verify the fit and aggregation of the obtained factors. Based on the information on correlation rules between students’ behavior and performance, precise performance management strategies for students are proposed, and targeted management methods for innovation and entrepreneurship education are formulated. The results indicate that the level of interest correlates with the enrollment grades of A and D, which are respectively 0.569 and 0.403. The degree of support, confidence Xing, and the degree of interest of the student’s enrollment grades of D and graduation grades of A are 0.234, 0.535, and 0.694. It shows that the student’s grades have a strong correlation with their interest in learning, and the education management should pay attention to the cultivation of the student’s interest in learning.