Material Diffusion Recommendation Algorithm Based on Tag Attributes
In order to solve the problem of data sparsity in recommendation system,a material diffusion recommendation algorithm based on tag attributes was proposed by using tag information between users and products.By constructing a user product tag bipartite graph network and a product user tag bipartite graph network,the recommended preference feature vectors for user tag attributes and for different types of products were calculated,and the resource allocation method in the material diffusion algorithm was im-proved accordingly.Comparative experimental results on the MovieLens movie rating data set show that the algorithm has improved to a certain extent in precision,recall,F1 coefficient and diversity.
万方数据
