Research on Personalized Recommendation System of Mixed Learning Resources Based on Data Mining

This paper deeply discusses the construction and optimization of personalized recommendation system of mixed learning resources based on data mining. This paper proposes an improved Collaborative average filtering (CAMF) algorithm, which can significantly improve the accuracy and adaptability of the recommendation system by introducing context information. This paper uses information gain as a metric to effectively identify and integrate the situational factors that have the greatest impact on learners' preferences, to make the recommendation process more accurate and personalized. This paper further develops the CAMF-CM hybrid recommendation algorithm, which combines the advantages of collaborative filtering and content to achieve comprehensive and in-depth recommendation of learning resources. The algorithm not only considers the historical behavior of learners, but also incorporates the intrinsic properties of learning resources and the immediate needs of learners in specific situations, thus finding a balance between diversity and accuracy. In the system simulation experiment, the recommended system in this paper shows excellent performance. Compared with traditional recommendation algorithms, CAMF-CM algorithm has achieved significant improvement in recommendation accuracy, recall rate and coverage rate. In addition, the efficient operation of the system and the friendliness of the user interface have also been praised by the experiment participants.