An Improved Collaborative Recommendation Method Based on Sna

Collaborative recommendation is widely used in e-commerce personalized service, but due to data sparsity and cold start, the existing method cannot give precise results. To improve the recommendation precision, this paper gives a new collaborative recommendation method based on SNA. The proposed method uses social network analysis( SNA) technical to analyze the trust between the users, expresses it as trust value to fill the users—items matrix, and the user similarity calculation has been improved. Finally, an experiment is used to verify the validation of the proposed method. It is proved that it can better solve the problem of data sparsity and cold start.