In this paper, we constructed a curriculum design ontology for grassland ecological study for the individual characteristics of learners, learning needs and the logical relationship between knowledge points, and extracted, fused and stored the knowledge mapping data of the discipline. After that, the structural relationship between knowledge points is explained, and the personalized learning path recommendation model is constructed based on the feature information of learners and learning resources in the learning path recommendation solving problem, with personalized learning path as the decision variable and based on the mapping relationship between learners and learning resources. Finally, the basic particle swarm optimization algorithm is improved from the perspectives of inertia weight and variation operator setting to solve the model. The MABPSO model has better convergence ability, stability and adaptability, and it can solve the personalized online learning resources recommendation problem better. In addition, the method in this paper utilizes the learner cognitive level function to calculate the learner’s mastery of the knowledge points, and then sets the generation rules of the learning paths, and finally generates the optimal learning paths and recommends them to the learners.
