Research On Collaborative Filtering Recommendation Algorithm Based On Social Relation And Matrix Completion

In recent years,with the rapid development of computer technology and the Internet,People have entered a new era of big data.The coming of the big data speeds up the expansion of data,the traditional information retrieval system and recommendation system cannot meet the demand of the users.Recommendation system has eased the problem that the searching results are not accurate enough from a certain extent,but it also exists some other problems under the background of the new big data.Collaborative filtering recommendation algorithm is one of the most classic recommendation algorithms,this paper made some improvement and innovation based on the problems of data sparseness and cold starting existed in the algorithm of collaborative filtering recommendation.It joined the ideas of social relation into the process of the algorithm of collaborative filtering recommendation and improved the way of the completion of matrix.The main work is as follows:On the one hand,the paper put the ideas of social relation into the process of the recommendation algorithm and put forward the CF-SR.The algorithm obtained the collection of each user's friends by the analysis of social relation,and then recommended for the target user according to the interests of his friends.This algorithm can solve the problem of cold starting from a certain extent,the paper verified which through the concrete experiment.On the other hand,complete the matrix conditionally and put forward the CFSRCC.The algorithm improved the way of the matrix completion further based on the CF-SR,it selected the items satisfied the certain conditions for completion,whichmade the matrix after the completion more accurate and reduced the redundancy of data.The algorithm can solve the problem of data sparseness to a certain extent and make the results more accurate.This paper analyzed the algorithm through the specific experiment,the experimental results showed that the improved algorithm has made the results of recommendation more accurate,at the same time,the algorithm is more efficient,and the value of the MAE and the RMSE is better.