With the further development of digital campus construction, the trend of modernisation of education is becoming more and more obvious, in this context will actively change the traditional way of education, but in the data processing stage of teaching materials, due to the lack of analysis of the attribute characteristics of different teaching materials, students are less satisfied with the combination of teaching resources in the actual teaching stage. Therefore, a research on the data processing method of ideological and political teaching materials of pharmacy specialty based on machine learning model is proposed. Under the constraint of fixed principles, the single ontology, multi ontology and mixed ontology of ideological and political teaching resources for pharmacy specialty were constructed. After preprocessing the teaching resource ontology of ideological and political teaching materials of pharmacy specialty with machine learning model, the attribute characteristics of different teaching materials are determined by grid search, and the variant individuals under the objective function are determined. Finally, the parameter combination with the highest score is determined by cross validation as the processing result of ideological and political teaching materials of pharmacy specialty. In the test results, the students gave good feedback to the corresponding teaching resource combination under the design method, obviously due to the control group.