利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

4 条记录 (耗时 0.037 秒)

    Automated objective-based feature improvement
    1.
    发明授权
    Automated objective-based feature improvement 有权
    自动基于目标的功能改进

    公开(公告)号：US08706733B1

    公开(公告)日：2014-04-22

    申请号：US13560874

    申请日：2012-07-27

    申请人： Zach Yeskel ,  David Andrew Huffaker ,  Rachel Ida Rosenthal Schutt ,  Andrew Stephen Tomkins ,  David Andrew Gibson ,  Abhijit Bose ,  Alexander Fabrikant ,  Makoto Uchida

    发明人： Zach Yeskel ,  David Andrew Huffaker ,  Rachel Ida Rosenthal Schutt ,  Andrew Stephen Tomkins ,  David Andrew Gibson ,  Abhijit Bose ,  Alexander Fabrikant ,  Makoto Uchida

    IPC分类号： G06F7/00 ,  G06F17/30

    CPC分类号： G06Q10/10

    摘要： A system and method for facilitating automatic objective-based feature improvement, the method including receiving a request to identify an optimal alternative for each of one or more features of a computer-implemented entity, determining an alternative feature range for each of the one or more features, the alternative feature range defining a range of possible alternative features available with respect to the feature, selecting one or more alternative features for each feature from the alternative feature range of the feature, generating a plurality of alternative sets, each including an alternative feature for at least one of the one or more features, selecting a plurality of user groups from a pool of users and assigning each user group of the plurality of users groups to one of the plurality of alternative sets based on user characteristics of the users and historical information regarding the interaction of the user with the computer-implemented entity.

    摘要翻译： 一种用于促进自动基于目标的特征改进的系统和方法，所述方法包括接收用于为计算机实现的实体的一个或多个特征中的每一个标识最佳替代的请求，确定所述一个或多个特征中的每一个的替代特征范围 特征，所述替代特征范围定义相对于所述特征可用的可能替代特征的范围，从所述特征的替代特征范围中为每个特征选择一个或多个替代特征，生成多个备选集合，每个替代集合包括替代特征 对于所述一个或多个特征中的至少一个，从用户池中选择多个用户组，并且基于所述用户的用户特征和所述多个用户组的历史，将所述多个用户组中的每个用户组分配给所述多个替代集合中的一个 关于用户与计算机实现的实体的交互的信息。

    Pre-processing of social network structures for fast discovery of cohesive groups
    3.
    发明授权
    Pre-processing of social network structures for fast discovery of cohesive groups 有权
    预处理社交网络结构，快速发现凝聚力群体

    公开(公告)号：US08909646B1

    公开(公告)日：2014-12-09

    申请号：US13731505

    申请日：2012-12-31

    申请人： Alexander Fabrikant ,  Atish Das Sarma

    发明人： Alexander Fabrikant ,  Atish Das Sarma

    IPC分类号： G06F17/30

    CPC分类号： G06F17/30601 ,  G06Q50/01

    摘要： Aspects and implementations of the present disclosure are directed to methods and systems of pre-processing a social network structure for fast discovery of cohesive groups. In general, in some implementations, a data processing system identifies a cohesive user group in a social network for delivery of a tailored content item. Generally, the data processing system identifies an affinity criteria; generates a set of user identifiers having characteristics that satisfy the affinity criteria; and generates graphs of users with at least one direct or indirect social network user connection with other user identifiers in the graph. The data processing system returns or stores the graph on computer readable media for later use. A graph may be generated with edges representing connections between user identifiers; edges may be weighted for the number and lengths of connection paths for indirect connections, and for similarities between users.

    摘要翻译： 本公开的方面和实现涉及预处理社交网络结构以快速发现粘性组的方法和系统。 通常，在一些实施方式中，数据处理系统识别社交网络中用于传送定制内容项目的一致性用户组。 通常，数据处理系统识别亲和度标准; 生成具有满足亲和度准则的特征的一组用户标识符; 并且生成具有与图中的其他用户标识符的至少一个直接或间接的社交网络用户连接的用户的图形。 数据处理系统将图形返回或存储在计算机可读介质上供以后使用。 可以生成具有表示用户标识符之间的连接的边的图; 边缘可以对于间接连接的连接路径的数量和长度以及用户之间的相似性进行加权。