Due to the development of advanced information technology such as artificial intelligence, the traditional marketing profession is being transformed and upgraded in the direction of intelligent higher vocational marketing, and the requirements of marketing positions on the knowledge, quality and ability of practitioners have changed. The article analyzes students’ cell phone online behavior in different classrooms based on DBSCAN clustering algorithm by collecting students’ campus network usage data, according to which the results can provide an effective basis for school management. By introducing the Interpretive Structural Model (ISM) and analyzing the interrelationships between courses, the article proposes a course cluster division scheme for marketing majors, which provides methodological support for the division of clusters in the construction of course clusters for professional teachers, as well as the selection and organization of the courses within the clusters. Finally, investigate the differential judgment of students from different places of origin about the influence of teaching environment, teacher quality, teaching process, teaching tools and resources on the teaching effect of marketing courses, the data show that the influence factors of marketing course teaching have obvious differences in the influence of the teaching effect of the course, improve the ability of professional teachers to educate people, optimize the teaching process of the marketing course, and deepen the reform of classroom teaching.