Friend recommendation method based on single-source SimRank exact solution

Abstract

The invention discloses a friend recommendation method based on a single-source SimRank exact solution. The friend recommendation method comprises the following steps of: converting a target user, users and a relationship among the users into a graph structure G; calculating personalized Page ranks of a source node vi relative to all nodes on the graph, forming a personalized Page ranking vector pi<right arrow>, calculating a no-more-meeting probability of all nodes on the graph structure G, forming a no-more-meeting probability matrix D<^>, calculating the SimRank similarity of the sourcenode vi according to an n-dimensional vector pi<right arrow> and the no-more-meeting probability matrix D<^> to obtain an n-dimensional vector S<right arrow>, repeatedly executing L rounds of calculation of the SimRank similarity, and updating the n-dimensional vector S<right arrow>; and finding out a node corresponding to the t dimension with the maximum value in all dimensions of an n-dimensional vector S<right arrow L>, and recommending the users corresponding to the t nodes to the target user as a result. The friend recommendation method based on the single-source SimRank exact solutioncan guarantee to obtain the exact solution of the single-source SimRank similarity on a large-scale user group within valid time, and the quality and effect of a friend recommendation function are improved.

Description

technical field [0001] The invention relates to a recommendation method, in particular to a friend recommendation method based on the exact solution of single-source SimRank. Background technique [0002] With the accelerated penetration of China's mobile Internet among the people, there has been a large-scale growth of mobile social users. The "2019 China Mobile Social Industry Special Report" released by iiMedia Consulting shows that the number of China Mobile social users reached 737 million in 2018, and it is expected to continue to grow steadily in the next two years. In 2020, it is expected to exceed 800 million. The group highlights the huge market space and also stimulates the emergence of various social software. According to the functions of social software, mainstream software can be roughly divided into instant messaging social applications represented by WeChat and QQ; media social applications represented by Weibo; interest social applications represented by D...

AI Technical Summary

Problems solved by technology

With the rapid expansion of the user scale, the calculation time of SimRank similarity also increases. In the face of large-scale user groups (such as Twitter, Weibo and other social applications with tens of millions of users), the existing methods cannot be used effectively. Accurately calculate the SimRank similarity between the user to be recommended and all users within a short period of time, only to sacrifice a certain degree of accuracy to obtain an estimated value of SimRank similarity, which greatly affects the quality and effect of the friend recommendation function

Images