In the context of the digital era, the association between user needs and learning objects is established by combining the characteristics of the collection of learning objects. Apply the angle cosine to calculate the distance between two vectors, and the distance is the personalized history knowledge course structure value. By virtue of the association matrix between history course learning resources and concept sequences, personalized history course learning content is generated for learning objects. The results show that compared with other methods, the F1 value of this paper's method is improved by 5.67% and 3.41% on average, and the largest percentage of grades above 80 in units 1-8 of the history course. It indicates that the recommendation algorithm to generate personalized history course content to accomplish students' course learning objectives.