A Recommendation of Pension Service Based on Trusted Network

The recommended modules have been applied to all walks of life, and the recommendation of the pension service less and not accurate at present. This paper proposes a recommendation of the pension based on trusted network, which aims to recommend reliable and appropriate services to users. The research of this paper is based on the platform of institutional pension service. First, create user portraits and service portraits, characterize users and services, and use k-means clustering algorithms to cluster users and services together. Secondly, build a trusted network. Establish a user trust model and use random walk algorithm to get trusted service. In addition, the impact of dynamic addition and deletion of users on trusted networks has also been dealt with accordingly; The experiment used Jingdong review data to verify the experimental results by analyzing user behavior. Finally, the experimental results were obtained through the analysis of user behaviors and logs. The experimental results have shown that, compared with the traditional service recommendation method, the recommendation of service based on the trusted network can obtain higher response rate of users.