Exploration of a personalised ideological and political education model based on a recommendation algorithm

In response to the problems of "efficiency bottleneck" and insufficient "personalization" in ideological and political education in universities under the background of the information age, this project innovatively integrates data mining and machine learning methods to establish a deep level user behavior analysis and recommendation method system based on big data. It fully collects students’ online learning trajectories and social media data, and accurately captures individual interests and needs through refined portrait construction and cluster analysis. On this basis, the optimal collaborative screening technique is adopted to personalize the matching and recommendation of different types of educational resources, and provide students with personalized learning paths. The experimental results show that this teaching mode can fully mobilize students’ learning enthusiasm, improve the pertinence and effectiveness of teaching, and the average academic performance of students is between 3-7 points, which tests its overall effect. The research results of this project will further expand the application scope of educational recommendation algorithms, provide new ideas for optimizing resource allocation and improving the quality of personalized education, and provide important technical and theoretical support for promoting the reform and development of ideological and political education in Chinese universities.