Research on the application of big data learning recommendation model driven by knowledge graph algorithm in educational information platform optimization

Driven by the global wave of digitalization, the education industry is undergoing profound changes, and the migration of traditional classroom teaching to online education is accelerating, giving rise to a large number of education information platforms. However, the problem of accurate matching of massive educational resources with learners' individual needs to be solved urgently. This study focuses on the application of knowledge graph algorithm and big data analysis technology in the education information platform, aiming to build an intelligent learning recommendation system. Firstly, a comprehensive knowledge graph covering multidisciplinary knowledge points is constructed, and a multi-level resource network is formed by mining the logical relationship of knowledge points. Then, combined with big data analysis of user behavior habits and learning preferences, a dynamic learning path recommendation algorithm was developed, and the recommendation list was adjusted in real time according to individual differences, and a personalized learning plan was generated. Experiments show that the optimized platform increases the average daily online time of users by 30 % and the course completion rate by >40 %, which significantly enhances user stickiness and learning effect. The research confirms that the system can improve student learning outcomes, stimulate engagement, and optimize the overall learning experience by accurately matching resources, providing a data-driven personalized teaching solution for the field of educational technology, and has practical value for promoting the development of intelligent education.