Recommending Users to a Social Networking System User Based on Relevance and Similarity Between Users
    1.
    发明申请
    Recommending Users to a Social Networking System User Based on Relevance and Similarity Between Users 审中-公开
    推荐用户基于用户之间的相关性和相似性的社交网络系统用户

    公开(公告)号:US20160147810A1

    公开(公告)日:2016-05-26

    申请号:US14553639

    申请日:2014-11-25

    Applicant: Facebook, Inc.

    CPC classification number: H04L65/403 G06F16/9535

    Abstract: A social networking system allows users to establish connections with other users to receive content provided to the social networking system by the other user. The social networking system may recommend an additional user to the user in response to the user establishing a connection with a target user. To recommend an additional user, the social networking system determines a relevance of the additional user to the target user relative to the relevance of other candidate users to the target user. The relevance may be based on a measure of similarity between an additional user and the target user, such as a covariance in the overlap of users connected to both the additional user and to the target user.

    Abstract translation: 社交网络系统允许用户与其他用户建立连接,以接收其他用户提供给社交网络系统的内容。 响应于用户建立与目标用户的连接,社交网络系统可向用户推荐额外的用户。 为了推荐额外的用户,社交网络系统相对于其他候选用户与目标用户的相关性来确定附加用户与目标用户的相关性。 相关性可以基于附加用户和目标用户之间的相似度的度量,例如连接到附加用户和目标用户的用户的重叠中的协方差。

Patent Agency Ranking