| With the application of social software,the link between people becomes more close,and the impact of social networks on personal behavior is also obvious.In this context,the traditional recommendation model using user ratings has been unable to meet the needs of users.In order to recommend for each user accurately,it's important to establish a recommendation model based on social network environment.In this paper,we choose the microblog trust relationship as the research object,and study the application of user trust relationship in the recommendation system.First of all,based on the research of social network recommendation model,this paper selects the microblog user trust relationship as the research object,and puts forward the method of calculating the trust degree of users based on the time effect by taking the existing model as the starting point.Then,the paper studies the recommendation model in the binary trust network.the paper integrate the similarity between the trusted users,and decompose the user score matrix based on the probability matrix decomposition model.In order to establish Enhance_MF model,the paper introduce the binary trust network into the probability matrix decomposition.Secondly,a recommendation model based on microblogging trust relationship(Blog_TrustMF)is established.In order to realizes the application of relationship between microblogging trust in Recommendation,the model takes the user influence factor as the influence weight among users based on the matrix decomposition of user trust relationship.Finally,we choose the RMSE as the evaluation index,and give the experimental evaluations and analysis of the algorithms proposed in this paper based on the Epinions data set and the Tencent microblog data set.The experimental results show that the proposed model has a certain improvement in the recommended accuracy compared with the similar model,and it enriches the existing social recommend model and provides a reference for other social networks. |
PDF Full Text Request