公开(公告)号：US09053129B1

公开(公告)日：2015-06-09

申请号：US13803180

申请日：2013-03-14

Applicant: Google Inc.

Inventor： Michele Covell ,  Shumeet Baluja

IPC: G06F17/30

CPC classification number: G06Q30/0241 ,  G06F17/3089

Abstract: Methods, systems, and apparatus for determining content item quality based on content item presentation data are presented. In one aspect, a method includes storing a first count of occurrences of a pair of content items being presented; storing a second count of second occurrences of a first content item of the pair being interacted with when it was presented in a lesser position in a resource relative to a second content item of the pair when the second content item of the pair was not interacted with; determining that the first count meets a first threshold; determining that a ratio of the second count to the first count meets a second threshold; and storing correlation data for the second content item based on the ratio.

Abstract translation: 提出了基于内容项呈现数据确定内容项目质量的方法，系统和装置。 一方面，一种方法包括存储正在呈现的一对内容项的出现的第一计数; 存储当所述对中的第二内容项不与所述对的第二内容项相互作用时相对于所述对中的所述第二内容项相对于所述对中的第二内容项的第二次出现的第二次出现， ; 确定所述第一计数满足第一阈值; 确定所述第二计数与所述第一计数的比率满足第二阈值; 以及基于所述比率存储所述第二内容项目的相关数据。

公开(公告)号：US20150039418A1

公开(公告)日：2015-02-05

申请号：US13959247

申请日：2013-08-05

Applicant: Google Inc.

Inventor： Shumeet Baluja ,  Joshua J. Sacks ,  Michael Chu ,  Zhengrong Ji ,  Michele Covell

IPC: G06Q30/02

CPC classification number: G06Q30/0246

Abstract: Methods, systems, and apparatus for content item distribution based on user interactions. In one aspect, a method includes identifying a set of conversion events associated with a content item provider, each conversion event having a corresponding device identifier; identifying, for each conversion event, a number of pre-conversion interactions that i) occurred prior to the conversion event, and ii) are associated with a device identifier that corresponds to the conversion event; generating, for the content item provider, a conversion profile specifying a portion of the conversion events that are associated with different pre-conversion interaction values; and generating, based on the conversion profile, a bid profile specifying different bid values for different pre-conversion interaction values, a difference between the different bid values being based on a difference between the portion of the conversion events that are associated with the different pre-conversion interaction values corresponding to the different bid values.

Abstract translation: 基于用户交互的内容分发方法，系统和设备。 在一个方面，一种方法包括识别与内容项提供者相关联的一组转换事件，每个转换事件具有对应的设备标识符; 为每个转换事件识别I）在转换事件之前发生的转换前交互的数量，以及ii）与对应于转换事件的设备标识符相关联; 为所述内容项提供者生成指定与不同的转换前交互值相关联的转换事件的一部分的转换简档; 以及基于所述转换简档，生成针对不同的转换前交互值指定不同的出价值的出价简档，所述不同的出价值之间的差异是基于与所述不同的预转换相关联的所述转换事件的所述部分之间的差异 - 转换相互作用值对应于不同的出价值。

公开(公告)号：US08886585B1

公开(公告)日：2014-11-11

申请号：US14070463

申请日：2013-11-01

Applicant: Google Inc.

Inventor： Rohan Seth ,  Joshua J. Sacks ,  Michele Covell ,  Michael Chu ,  Shumeet Baluja ,  David Marwood

IPC: G06N5/02 ,  G06F7/00 ,  G06Q30/02

CPC classification number: G06Q30/0251 ,  G06F9/4411 ,  G06F17/30864 ,  G06F17/30867 ,  G06N99/005 ,  G06Q30/02

Abstract: A method and computer program product for providing content to a user or computing device is disclosed. A plurality of operating modes for a computing device are identified based on usage information generated for the computing device over one or more periods of time. A request for content to display at the computing device is received, and the computing device is determined to be operating in at least one of the operating modes. User targeting information for the at least one of the operating modes is determined based on at least a portion of the usage information, and content for display at the computing device is provided based on the targeting information for the at least one of the operating modes.

Abstract translation: 公开了一种用于向用户或计算设备提供内容的方法和计算机程序产品。 基于在一个或多个时间段上为计算设备生成的使用信息来识别用于计算设备的多个操作模式。 接收到在计算设备上显示内容的请求，并且确定计算设备在至少一个操作模式下操作。 基于所述使用信息的至少一部分来确定用于所述至少一个所述操作模式的用户定位信息，并且基于所述至少一个所述操作模式的所述目标信息来提供在所述计算设备处显示的内容。

公开(公告)号：US20140115010A1

公开(公告)日：2014-04-24

申请号：US13778361

申请日：2013-02-27

Applicant: Google Inc.

Inventor： Rohan Seth ,  Shumeet Baluja ,  Michele Covell

IPC: G06F17/30

CPC classification number: G06F16/9024 ,  G06F16/95 ,  G06Q50/01

Abstract: Methods, and systems, including computer programs encoded on computer-readable storage mediums, including a method for providing a graph that includes entity nodes, label nodes and weighted connecting edges. The method comprises computing an aggregated incoming between-entity edge weight for the entity nodes. When there are positively-weighted incoming between-entity edges into the entity node, the method comprises replacing each of the between-entity edge weights by a pre-normalized between-entity edge weights. The method comprises computing an aggregated from-label weight for the entity node. When there are positively-weighted from-label node edges, the method comprises replacing the corresponding label weights by pre-normalized from-label weights. The method comprises determining influence values for a first, second and third influence factors, where the influence factors have values that sum to one. The method further comprises using the pre-normalized weights and influence factors as a set of linear constraints to determine final label weightings for the entity nodes.

Abstract translation: 方法和系统，包括在计算机可读存储介质上编码的计算机程序，包括用于提供包括实体节点，标签节点和加权连接边缘的图形的方法。 该方法包括计算实体节点的聚合进入的实体间边缘权重。 当在实体节点之间存在正加权的进入实体边缘之间时，该方法包括通过预定义的实体之间的权重来替换每个实体之间的权重。 该方法包括计算实体节点的聚合的标签权重。 当具有正加权的标签节点边缘时，该方法包括通过预标准化的标签权重来替换对应的标签权重。 该方法包括确定第一，第二和第三影响因素的影响值，其中影响因素具有值为1的值。 该方法还包括使用预标准化权重和影响因子作为一组线性约束来确定实体节点的最终标签权重。

公开(公告)号：US20190279289A1

公开(公告)日：2019-09-12

申请号：US15346907

申请日：2016-11-09

Applicant: Google Inc.

Inventor： Shumeet Baluja ,  Joshua J. Sacks ,  Michael Chu ,  Zhengrong Ji ,  Michele Covell

IPC: G06Q30/08 ,  H04L12/26

Abstract: Methods, systems, and apparatus for content item distribution based on user interactions. In one aspect, a method includes identifying a set of conversion events associated with a content item provider, each conversion event having a corresponding device identifier; identifying, for each conversion event, a number of pre-conversion interactions that i) occurred prior to the conversion event, and ii) are associated with a device identifier that corresponds to the conversion event; generating, for the content item provider, a conversion profile specifying a portion of the conversion events that are associated with different pre-conversion interaction values; and generating, based on the conversion profile, a bid profile specifying different bid values for different pre-conversion interaction values, a difference between the different bid values being based on a difference between the portion of the conversion events that are associated with the different pre-conversion interaction values corresponding to the different bid values.

公开(公告)号：US10282430B1

公开(公告)日：2019-05-07

申请号：US14804361

申请日：2015-07-21

Applicant: Google Inc.

Inventor： Rohan Seth ,  Joshua J. Sacks ,  Michele Covell ,  Michael Chu ,  Shumeet Baluja ,  David Marwood

IPC: G06F17/30 ,  G06T1/00

Abstract: Methods, systems, and computer program products are provided for determining content items for delivery to users based on imagery. One example method includes determining a location of a user or a location associated with a query submitted by the user, identifying imagery associated with the determined location, and determining one or more content items for delivery to the user based at least in part on the imagery.

公开(公告)号：US09547696B2

公开(公告)日：2017-01-17

申请号：US14845825

申请日：2015-09-04

Applicant: GOOGLE INC.

Inventor： Rohan Seth ,  Shumeet Baluja ,  Michael Jesse Chu ,  Joshua J. Sacks ,  David Marwood ,  Michele Covell

IPC: G06F17/30

CPC classification number: G06F17/3053 ,  G06F17/30241 ,  G06F17/30528 ,  G06F17/30554 ,  G06F17/3087

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer-readable storage medium, including a method for ranking content using location-based query log analysis. The method comprises: identifying a region defining an area of interest including identifying a plurality of content items that are associated with the region; evaluating query logs associated with users that submitted queries associated with the region to determine a ranking associated with the plurality of content items; receiving a request for content associated with the region; and providing one or more of the content items based at least in part on the ranking.

Abstract translation: 方法，系统和装置，包括在计算机可读存储介质上编码的计算机程序，包括使用基于位置的查询日志分析来对内容进行排序的方法。 该方法包括：识别定义感兴趣区域的区域，包括识别与该区域相关联的多个内容项目; 评估与提交与所述区域相关联的查询的用户相关联的查询日志以确定与所述多个内容项目相关联的排名; 接收与该地区相关联的内容的请求; 以及至少部分地基于排名来提供一个或多个内容项目。

公开(公告)号：US09384571B1

公开(公告)日：2016-07-05

申请号：US14024330

申请日：2013-09-11

Applicant: Google Inc.

Inventor： Michele Covell ,  Shumeet Baluja

IPC: G06F7/00 ,  G06F17/30 ,  G06T11/20

CPC classification number: G06T11/206 ,  G06F17/30365 ,  G06F17/30368 ,  G06F17/30958 ,  G06Q10/10 ,  G06Q50/01

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for updating graphs. Labels associated with nodes of a graph are identified, including designators describing an attribute associated with a given node. The graph is provided, wherein labels have been assigned to each node in the graph. An initial set of weights for the labels are assigned reflecting a magnitude of a contribution of an associated label to a characterization of a respective node. A portion of the labels are assigned based on a propagation from other nodes. A change is identified in the graph that, when propagated, will affect other nodes. Sparse matrices, generated to describe the change, contain nonzero entries only in rows wherein connection weights and/or labels have changed. A new graph is generated using the sparse matrices without having to recalculate weights for other nodes not affected by the change.

Abstract translation: 方法，系统和装置包括在计算机可读存储介质上编码的计算机程序，包括用于更新图形的方法。 识别与图的节点相关联的标签，包括描述与给定节点相关联的属性的指示符。 提供了图表，其中标签已被分配给图中的每个节点。 分配标签的初始权重集合反映相关标签对相应节点的表征的贡献的大小。 基于来自其他节点的传播，分配了一部分标签。 在图中确定了一个变化，当传播时，会影响其他节点。 生成以描述更改的稀疏矩阵仅包含连接权重和/或标签已更改的行中的非零条目。 使用稀疏矩阵生成新图，而不必重新计算不受更改影响的其他节点的权重。

公开(公告)号：US09137529B1

公开(公告)日：2015-09-15

申请号：US14208352

申请日：2014-03-13

Applicant: Google Inc.

Inventor： Michele Covell ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja ,  Vivek Kwatra

IPC: G06F15/18 ,  H04N19/105 ,  H04N19/17 ,  H04N19/189

CPC classification number: G06K9/6202 ,  G06F17/30247 ,  G06K9/00684 ,  G06K9/4604 ,  G06K9/6267 ,  G06K9/6292 ,  G06K9/66 ,  G06T1/0021 ,  H04N19/105 ,  H04N19/17 ,  H04N19/189

Abstract: An exemplar dictionary is built from exemplars of digital content for determining predictor blocks for encoding and decoding digital content. The exemplar dictionary organizes the exemplars as clusters of similar exemplars. Each cluster is mapped to a label. Machine learning techniques are used to generate a prediction model for predicting a label for an exemplar. The prediction model can be a hashing function that generates a hash key corresponding to the label for an exemplar. The prediction model learns from a training set based on the mapping from clusters to labels. A new mapping is obtained that improves a measure of association between clusters and labels. The new mapping is used to generate a new prediction model. This process is repeated in order to iteratively refine the machine learning modes generated.

Abstract translation: 由数字内容的示例构建示范字典，用于确定用于对数字内容进行编码和解码的预测器块。 示范字典将样本组织成类似样本的集群。 每个集群映射到一个标签。 机器学习技术用于生成用于预测样本的标签的预测模型。 预测模型可以是哈希函数，其生成与样本的标签相对应的散列密钥。 基于从集群到标签的映射，预测模型从训练集学习。 获得了一种新的映射，改进了集群和标签之间的关联度量。 新映射用于生成新的预测模型。 重复该过程以便迭代地改进所生成的机器学习模式。

公开(公告)号：US20150169725A1

公开(公告)日：2015-06-18

申请号：US14264411

申请日：2014-04-29

Applicant: GOOGLE INC.

Inventor： Yushi Jing ,  Michele Covell ,  Stephen C. Holiday

IPC: G06F17/30

CPC classification number: G06F17/30598 ,  G06F17/30256 ,  G06F17/3028 ,  G06F17/3053

Abstract: Aspects of the subject matter described herein relate to functions used for retrieving image results based on search queries. More specifically, image search queries can be pre-grouped or classified based on visual and semantic similarity. For example, a pairwise image similarity value for a pair of queries can be computed based on one or more of the sum of all of the overlapping the image results, the sum of the image distances between all of the pairs of images in the image results, and the rank of each of the images in the image results. The pairwise image similarity values can then be used to generate image query clusters. Each image query clusters can include a set of queries with high pairwise image similarity values. In some examples, a distance function can be determined for each image query cluster. This data can be used to provide image results.

Abstract translation: 本文描述的主题的方面涉及用于基于搜索查询来检索图像结果的功能。 更具体地，可以基于视觉和语义相似性对图像搜索查询进行预分组或分类。 例如，可以基于图像结果重叠的全部和之和中的一个或多个来计算一对查询的成对图像相似度值，图像结果中所有图像对之间的图像距离之和 ，以及图像中每个图像的等级。 然后可以使用成对图像相似度值来生成图像查询簇。 每个图像查询群集可以包括具有高成对图像相似度值的一组查询。 在一些示例中，可以为每个图像查询簇确定距离函数。 该数据可用于提供图像结果。