公开(公告)号：US20090133048A1

公开(公告)日：2009-05-21

申请号：US12120217

申请日：2008-05-13

申请人： Simon J. Gibbs ,  Bottyan Nemeth ,  Priyang Rathod ,  Anugeetha Kunjithapatham ,  Mithun Sheshagiri ,  Phuong Nguyen

发明人： Simon J. Gibbs ,  Bottyan Nemeth ,  Priyang Rathod ,  Anugeetha Kunjithapatham ,  Mithun Sheshagiri ,  Phuong Nguyen

IPC分类号： H04H60/32

CPC分类号： G06F17/30038

摘要： System and method for automatically rating the content of video media based on video operations performed on a media device and in reference to a plurality of rating rules are provided. Usage of the media device is continuously monitored and user actions with respect to operating the video media on the media device are automatically logged. Each rating rule includes a device usage pattern with respect to operating videos on the media device and a rating action indicating adjustments to content ratings of the videos based upon characteristics described by the device usage pattern that are inferred from the recorded user inputted video control operations. When the device usage pattern of a rating rule is inferred from one or more user actions operating a piece of video media directly on the media device, the content rating of the piece of video media is adjusted based on the rating rule.

摘要翻译： 提供了基于在媒体设备上执行的视频操作并参考多个评级规则来自动对视频媒体的内容进行评级的系统和方法。 持续监控媒体设备的使用情况，并自动记录用户在媒体设备上操作视频媒体的操作。 每个评级规则包括关于媒体设备上的操作视频的设备使用模式，以及基于从记录的用户输入的视频控制操作推断出的设备使用模式所描述的特性的指示对视频的内容评级进行调整的评级动作。 当通过直接在媒体设备上操作一段视频媒体的一个或多个用户动作来推断评级规则的设备使用模式时，基于评级规则来调整该片视频媒体的内容分级。

公开(公告)号：US20090132520A1

公开(公告)日：2009-05-21

申请号：US12120211

申请日：2008-05-13

申请人： Bottyan Nemeth ,  Simon J. Gibbs ,  Mithun Sheshagiri ,  Priyang Rathod

发明人： Bottyan Nemeth ,  Simon J. Gibbs ,  Mithun Sheshagiri ,  Priyang Rathod

IPC分类号： G06F17/30

CPC分类号： G06F17/30038

摘要： Various methods for combining ClipRank and Collaborative Filtering are provided. According to one embodiment, the ClipRank weights associated with a plurality of pieces of media content are calculated based on the relationships among the plurality of pieces of media content and a plurality of users. Those pieces having ClipRank weights greater than or equal to a predefined weight threshold are selected from the plurality of pieces of media content to obtain a plurality of selected pieces of media content. Collaborative Filtering is then performed on the plurality of selected pieces of media content and the plurality of users. According to another embodiment, Collaborative Filtering on a plurality of pieces of media content and a plurality of users is performed for one of the plurality of users. Personalized ClipRank weights associated with the plurality of pieces of media content is calculated for the user based on Collaborative Filtering ratings obtained for the plurality of pieces of media content for the user.

摘要翻译： 提供了组合ClipRank和协同过滤的各种方法。 根据一个实施例，基于多个媒体内容和多个用户之间的关系来计算与多个媒体内容相关联的ClipRank权重。 从多个媒体内容中选择具有大于或等于预定权重阈值的ClipRank权重的片段，以获得多个所选择的媒体内容。 然后对多个选定的媒体内容和多个用户执行协作过滤。 根据另一个实施例，对多个用户中的一个执行对多个媒体内容和多个用户的协作过滤。 基于针对用户的多个媒体内容获得的协作过滤等级，为用户计算与多个媒体内容相关联的个性化ClipRank权重。

公开(公告)号：US20090307296A1

公开(公告)日：2009-12-10

申请号：US12133304

申请日：2008-06-04

申请人： Simon J. GIBBS ,  Bottyan Nemeth

发明人： Simon J. GIBBS ,  Bottyan Nemeth

IPC分类号： G06F15/16 ,  G06F17/16

CPC分类号： G06F17/30029 ,  G06F17/30271 ,  G06F17/30867 ,  G06Q30/02

摘要： System and method for performing Collaborative Filtering while preserving complete user anonymity are provided. Each of a group of client devices sends a rating vector anonymously to a server. The cells in each rating vector correspond to a set of items, and selected cells have ratings provided by the user associated with the corresponding client device for the corresponding items. The server aggregates all the rating vectors into a rating matrix, and factorizes the rating matrix into a user feature matrix and an item feature matrix through approximation, such that the rating matrix equals the product of the user feature matrix and the item feature matrix. The item feature matrix is sent to the client devices. Each of the client devices calculates its own user feature vector based on its rating vector and the item feature matrix, and provides personalized recommendations on selected items based on the client's user feature vector and the item feature matrix.

摘要翻译： 提供用于执行协同过滤同时保持完整的用户匿名的系统和方法。 一组客户端设备中的每一个将评级向量匿名发送到服务器。 每个评分矢量中的单元对应于一组项目，并且所选择的单元格由与相应客户端设备相关联的用户为相应项目提供的评级。 服务器将所有评级向量聚合成评级矩阵，并通过近似将评级矩阵分解为用户特征矩阵和项目特征矩阵，使得评级矩阵等于用户特征矩阵和项目特征矩阵的乘积。 项目特征矩阵被发送到客户端设备。 每个客户设备基于其评级向量和项目特征矩阵来计算其自己的用户特征向量，并且基于客户端的用户特征向量和项目特征矩阵提供关于所选项目的个性化建议。