Intelligent Personalized Recommendation Method Based on Optimized Collaborative Filtering Algorithm in Primary and Secondary Education Resource System

Abstract

The Internet has driven the development of online education, and the vast system of educational resources has put forward higher requirements for personalized recommendation systems. In response to this issue, this study proposes a personalized recommendation system on the ground of optimized collaborative filtering algorithms. Due to the strong interaction between collaborative filtering algorithms and users, they are often used in recommendation models. However, its defects such as cold start can weaken the performance of the model. This study introduces content recommendation algorithms to address this phenomenon. A hybrid recommendation model on the ground of the two algorithms can effectively achieve personalized recommendations. Meanwhile, this study focuses on the key modules in the overall model and utilizes standardization and dimensionality reduction operations to further reduce the computational burden on the system. Finally, to verify the reliability of the model, the study compared it with other models. The experimental results showed that the accuracy of the mixed recommendation model was 2.68% higher than that of the utility recommendation model and the rule recommendation model, respectively, and 7.99%. Therefore, the personalized recommendation model on the ground of optimized collaborative filtering algorithm proposed in the study is effective.