Students’ ideology is mainly developed in the college period, so the ideological and political education work in colleges and universities should be strengthened in order to help students improve their ideological and moral qualities. In this paper, after the introduction of data mining, through the comparison of two algorithms of COBWEB clustering and K-means clustering based on the model method, we chose to use the K-means clustering method to carry out the learning characteristics mining clustering for the students of logistics management majoring in the school of management of a university in Zhengzhou City, Henan Province, and 220 students were selected as the research object. The students were divided into four categories for integration from six dimensions: learning foundation, learning interest, independent learning ability, collaborative learning ability, beginning of Civics learning, and advanced Civics learning. and. then, the purpose of students’ learning Civics education was studied in terms of learning motivation. In all dimensions, students in the first category did well, and 135 students in each category aimed to improve their academic qualifications while learning Civics courses. The innovative path for the content of Civic and Political Education is proposed based on the learning characteristics of the students. 180 students were selected as examples of logistics management class students, divided into experimental and comparison groups, and explored in terms of cultural self-confidence level, academic level, and motivation for learning using the difference analysis method. It was found that after applying the proposed content to the Civic and Political Education classroom, the experimental group and the comparison group showed significant improvement in all dimensions except for learning motivation. It indicates that the content innovation strategy of Civic and Political Education proposed in this paper has a better impact on the teaching effect.