专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20110131160A1 Method and System for Generating A Linear Machine Learning Model for Predicting Online User Input Actions 有权
标题翻译：用于生成用于预测在线用户输入操作的线性机器学习模型的方法和系统
公开(公告)号：US20110131160A1
公开(公告)日：2011-06-02
申请号：US13018303
申请日：2011-01-31
申请人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
发明人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
IPC分类号： G06F15/18
CPC分类号： G06Q30/02
摘要： A method of targeting receives several granular events and preprocesses the received granular events thereby generating preprocessed data to facilitate construction of a model based on the granular events. The method generates a predictive model by using the preprocessed data. The predictive model is for determining a likelihood of a user action. The method trains the predictive model. A system for targeting includes granular events, a preprocessor for receiving the granular events, a model generator, and a model. The preprocessor has one or more modules for at least one of pruning, aggregation, clustering, and/or filtering. The model generator is for constructing a model based on the granular events, and the model is for determining a likelihood of a user action. The system of some embodiments further includes several users, a selector for selecting a particular set of users from among the several users, a trained model, and a scoring module.
摘要翻译：定向的方法接收几个粒度事件并预处理所接收的粒状事件，从而生成预处理的数据，以便于基于粒状事件构建模型。该方法通过使用预处理数据生成预测模型。预测模型用于确定用户动作的可能性。该方法训练预测模型。用于定位的系统包括粒状事件，用于接收粒度事件的预处理器，模型生成器和模型。预处理器具有一个或多个用于修剪，聚合，聚类和/或过滤中的至少一个的模块。模型生成器用于基于粒度事件构建模型，模型用于确定用户操作的可能性。一些实施例的系统还包括若干用户，用于从几个用户中选择特定用户组的选择器，训练模型和评分模块。

2. 发明授权

US07921069B2 Granular data for behavioral targeting using predictive models 有权
标题翻译：使用预测模型的行为定位的粒度数据
公开(公告)号：US07921069B2
公开(公告)日：2011-04-05
申请号：US11770413
申请日：2007-06-28
申请人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
发明人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
IPC分类号： G06F17/00 , G06N5/02
CPC分类号： G06Q30/02
摘要： A method of targeting receives several granular events and preprocesses the received granular events thereby generating preprocessed data to facilitate construction of a model based on the granular events. The method generates a predictive model by using the pre-processed data. The predictive model is for determining a likelihood of a user action. The method trains the predictive mode. A system for targeting includes granular events, a preprocessor for receiving the granular events, a model generator, and a model. The preprocessor has one or more modules for at least one of pruning, aggregation, clustering, and/or filtering. The model generator is for constructing a model based on the granular events, and the model is for determining a likelihood of a user action. The system of some embodiments further includes several users, a selector for selecting a particular set of users from among the several users, a trained model, and a scoring module.
摘要翻译：定向的方法接收几个粒度事件并预处理所接收的粒状事件，从而生成预处理的数据，以便于基于粒状事件构建模型。该方法通过使用预处理数据生成预测模型。预测模型用于确定用户动作的可能性。该方法训练预测模式。用于定位的系统包括粒状事件，用于接收粒度事件的预处理器，模型生成器和模型。预处理器具有一个或多个用于修剪，聚合，聚类和/或过滤中的至少一个的模块。模型生成器用于基于粒度事件构建模型，模型用于确定用户操作的可能性。一些实施例的系统还包括若干用户，用于从几个用户中选择特定用户组的选择器，训练模型和评分模块。

3. 发明授权

US08364627B2 Method and system for generating a linear machine learning model for predicting online user input actions 有权
标题翻译：用于生成用于预测在线用户输入动作的线性机器学习模型的方法和系统
公开(公告)号：US08364627B2
公开(公告)日：2013-01-29
申请号：US13018303
申请日：2011-01-31
申请人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
发明人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
IPC分类号： G06F17/00 , G06N5/02
CPC分类号： G06Q30/02
摘要： A method of targeting receives several granular events and preprocesses the received granular events thereby generating preprocessed data to facilitate construction of a model based on the granular events. The method generates a predictive model by using the preprocessed data. The predictive model is for determining a likelihood of a user action. The method trains the predictive model. A system for targeting includes granular events, a preprocessor for receiving the granular events, a model generator, and a model. The preprocessor has one or more modules for at least one of pruning, aggregation, clustering, and/or filtering. The model generator is for constructing a model based on the granular events, and the model is for determining a likelihood of a user action. The system of some embodiments further includes several users, a selector for selecting a particular set of users from among the several users, a trained model, and a scoring module.
摘要翻译：定向的方法接收几个粒度事件并预处理所接收的粒状事件，从而生成预处理的数据，以便于基于粒状事件构建模型。该方法通过使用预处理数据生成预测模型。预测模型用于确定用户动作的可能性。该方法训练预测模型。用于定位的系统包括粒状事件，用于接收粒度事件的预处理器，模型生成器和模型。预处理器具有一个或多个用于修剪，聚合，聚类和/或过滤中的至少一个的模块。模型生成器用于基于粒度事件构建模型，模型用于确定用户操作的可能性。一些实施例的系统还包括若干用户，用于从几个用户中选择特定用户组的选择器，训练模型和评分模块。

4. 发明申请

US20090006363A1 Granular Data for Behavioral Targeting 有权
标题翻译：行为定位的粒度数据
公开(公告)号：US20090006363A1
公开(公告)日：2009-01-01
申请号：US11770413
申请日：2007-06-28
申请人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
发明人： John Canny , Shi Zhong , Scott Gaffney , Chad Brower , Pavel Berkhin , George H. John
IPC分类号： G06F17/30
CPC分类号： G06Q30/02
摘要： A method of targeting receives several granular events and preprocesses the received granular events thereby generating preprocessed data to facilitate construction of a model based on the granular events. The method generates a predictive model by using the pre-processed data. The predictive model is for determining a likelihood of a user action. The method trains the predictive mode. A system for targeting includes granular events, a preprocessor for receiving the granular events, a model generator, and a model. The preprocessor has one or more modules for at least one of pruning, aggregation, clustering, and/or filtering. The model generator is for constructing a model based on the granular events, and the model is for determining a likelihood of a user action. The system of some embodiments further includes several users, a selector for selecting a particular set of users from among the several users, a trained model, and a scoring module.
摘要翻译：定向的方法接收几个粒度事件并预处理所接收的粒状事件，从而生成预处理的数据，以便于基于粒状事件构建模型。该方法通过使用预处理数据生成预测模型。预测模型用于确定用户动作的可能性。该方法训练预测模式。用于定位的系统包括粒状事件，用于接收粒度事件的预处理器，模型生成器和模型。预处理器具有一个或多个用于修剪，聚合，聚类和/或过滤中的至少一个的模块。模型生成器用于基于粒度事件构建模型，模型用于确定用户操作的可能性。一些实施例的系统还包括若干用户，用于从几个用户中选择特定用户组的选择器，训练模型和评分模块。

5. 发明申请

US20100179855A1 Large-Scale Behavioral Targeting for Advertising over a Network 有权
标题翻译：通过网络广告的大规模行为定位
公开(公告)号：US20100179855A1
公开(公告)日：2010-07-15
申请号：US12351749
申请日：2009-01-09
申请人： Ye Chen , Dmitry Pavlov , Pavel Berkhin , John Canny
发明人： Ye Chen , Dmitry Pavlov , Pavel Berkhin , John Canny
IPC分类号： G06Q10/00 , G06F17/30 , G06F7/08 , G06N5/02
CPC分类号： G06Q30/02 , G06Q10/063 , G06Q10/0637 , G06Q10/0639 , G06Q30/0201 , G06Q30/0202 , G06Q30/0206
摘要： A method and a system are provided for large-scale behavioral targeting for advertising over a network, such as the Internet. In one example, the system receives training data that is processed raw data of user behavior. The system generates selected features by performing feature selection on the training data. The system generates feature vectors from the selected features. The system initializes weights of a behavioral targeting model by scanning the feature vectors once. The system then updates the weights of the behavioral targeting model by scanning iteratively the feature vectors using a multiplicative recurrence.
摘要翻译：提供了一种方法和系统，用于通过网络（如Internet）进行广告的大规模行为定位。在一个示例中，系统接收处理用户行为的原始数据的训练数据。系统通过对训练数据执行特征选择来生成选定的特征。系统从所选特征生成特征向量。系统通过扫描特征向量一次来初始化行为定位模型的权重。然后，系统通过使用乘法复现来迭代地扫描特征向量来更新行为定向模型的权重。

6. 发明授权

US08150723B2 Large-scale behavioral targeting for advertising over a network 有权
标题翻译：通过网络进行广告的大规模行为定位
公开(公告)号：US08150723B2
公开(公告)日：2012-04-03
申请号：US12351749
申请日：2009-01-09
申请人： Ye Chen , Dmitry Pavlov , Pavel Berkhin , John Canny
发明人： Ye Chen , Dmitry Pavlov , Pavel Berkhin , John Canny
IPC分类号： G06Q10/00 , G06Q30/00
CPC分类号： G06Q30/02 , G06Q10/063 , G06Q10/0637 , G06Q10/0639 , G06Q30/0201 , G06Q30/0202 , G06Q30/0206
摘要： A method and a system are provided for large-scale behavioral targeting for advertising over a network, such as the Internet. In one example, the system receives training data that is processed raw data of user behavior. The system generates selected features by performing feature selection on the training data. The system generates feature vectors from the selected features. The system initializes weights of a behavioral targeting model by scanning the feature vectors once. The system then updates the weights of the behavioral targeting model by scanning iteratively the feature vectors using a multiplicative recurrence.
摘要翻译：提供了一种方法和系统，用于通过网络（如Internet）进行广告的大规模行为定位。在一个示例中，系统接收处理用户行为的原始数据的训练数据。系统通过对训练数据执行特征选择来生成选定的特征。系统从所选特征生成特征向量。系统通过扫描特征向量一次来初始化行为定位模型的权重。然后，系统通过使用乘法复现来迭代地扫描特征向量来更新行为定向模型的权重。

7. 发明授权

US5546314A Processor and method for developing a set of admissible fixture designs for a workpiece 失效
标题翻译：用于开发一组工件可容许夹具设计的处理器和方法
公开(公告)号：US5546314A
公开(公告)日：1996-08-13
申请号：US466112
申请日：1995-06-06
申请人： Randolph C. Brost , Kenneth Y. Goldberg , Aaron S. Wallack , John Canny
发明人： Randolph C. Brost , Kenneth Y. Goldberg , Aaron S. Wallack , John Canny
IPC分类号： B23Q3/10 , G06F19/00
CPC分类号： B23Q3/103
摘要： A fixture process and method is provided for developing a complete set of all admissible fixture designs for a workpiece which prevents the workpiece from translating or rotating. The fixture processor generates the set of all admissible designs based on geometric access constraints and expected applied forces on the workpiece. For instance, the fixture processor may generate a set of admissible fixture designs for first, second and third locators placed in an array of holes on a fixture plate and a translating clamp attached to the fixture plate for contacting the workpiece. In another instance, a fixture vice is used in which first, second, third and fourth locators are used and first and second fixture jaws are tightened to secure the workpiece. The fixture process also ranks the set of admissible fixture designs according to a predetermined quality metric so that the optimal fixture design for the desired purpose may be identified from the set of all admissible fixture designs.
摘要翻译：提供了一种固定工艺和方法，用于开发用于工件的一整套允许的夹具设计，防止工件平移或旋转。夹具处理器基于几何访问约束和对工件的预期施加力产生所有允许设计的集合。例如，固定装置处理器可以产生一组可容许的夹具设计，用于放置在固定板上的孔阵列中的第一，第二和第三定位器以及附接到固定板的用于接触工件的平移夹具。在另一种情况下，使用夹具，其中使用第一，第二，第三和第四定位器，并且紧固第一和第二固定夹爪以固定工件。固定过程还根据预定的质量度量对一组允许的夹具设计进行排列，从而可以从所有允许的夹具设计的集合中识别用于期望目的的最佳夹具设计。

8. 发明授权

US09760802B2 Probabilistic recommendation of an item 有权
公开(公告)号：US09760802B2
公开(公告)日：2017-09-12
申请号：US12694903
申请日：2010-01-27
申请人： Ye Chen , John Canny
发明人： Ye Chen , John Canny
IPC分类号： G06Q30/02 , G06Q10/06 , G06Q30/06 , G06K9/62
CPC分类号： G06K9/6226 , G06Q30/02 , G06Q30/0251 , G06Q30/0254 , G06Q30/0257 , G06Q30/0269 , G06Q30/0282
摘要： A clustering and recommendation machine determines that an item is included in a cluster of items. The machine accesses item data descriptive of the item. The machine accesses a vector that represents the cluster and calculates the likelihood that the item is included in the cluster, based on the item variable and the probability parameter. The machine determines that the item is included in the cluster, based on the likelihood. The machine also recommends an item to a potential buyer. The machine accesses behavior data that represents a first event type pertinent to a first cluster of items. The machine calculates a probability that a second event type pertaining to a second cluster of items will co-occur with the first event type. The machine identifies an item from the second cluster to be recommended and presents a recommendation of the item to the potential buyer.

9. 发明申请

US20100306161A1 CLICK THROUGH RATE PREDICTION USING A PROBABILISTIC LATENT VARIABLE MODEL 审中-公开
标题翻译：使用概率偏差变量模型点击通过率预测
公开(公告)号：US20100306161A1
公开(公告)日：2010-12-02
申请号：US12474668
申请日：2009-05-29
申请人： Ye Chen , Dmitry Pavlov , John Canny , Eren Manavoglu
发明人： Ye Chen , Dmitry Pavlov , John Canny , Eren Manavoglu
IPC分类号： G06N7/02 , G06Q30/00 , G06N5/02 , G06Q10/00 , G06F17/30 , G06F15/18
CPC分类号： G06N7/005 , G06Q30/02 , G06Q30/0242 , G06Q30/0254 , G06Q30/0256
摘要： Methods and systems are provided for predicting click through rate in connection with a particular user, keyword-based query, and advertisement using a probabilistic latent variable model. Click through rate may be predicted based on historical sponsored search activity information. Predicted click through rate may be used as a factor in determining advertisement rank.
摘要翻译：提供了方法和系统，用于使用概率潜变量模型来预测与特定用户相关联的点击率，基于关键字的查询和广告。点击率可以根据历史赞助搜索活动信息进行预测。预测点击率可用作确定广告排名的因素。

10. 发明申请

US20110184806A1 PROBABILISTIC RECOMMENDATION OF AN ITEM 有权
标题翻译：项目概况建议
公开(公告)号：US20110184806A1
公开(公告)日：2011-07-28
申请号：US12694903
申请日：2010-01-27
申请人： Ye Chen , John Canny
发明人： Ye Chen , John Canny
IPC分类号： G06Q30/00 , G06F3/14 , G06F15/18
CPC分类号： G06K9/6226 , G06Q30/02 , G06Q30/0251 , G06Q30/0254 , G06Q30/0257 , G06Q30/0269 , G06Q30/0282
摘要： A clustering and recommendation machine determines that an item is included in a cluster of items. The machine accesses item data descriptive of the item. The machine accesses a vector that represents the cluster and calculates the likelihood that the item is included in the cluster, based on the item variable and the probability parameter. The machine determines that the item is included in the cluster, based on the likelihood. The machine also recommends an item to a potential buyer. The machine accesses behavior data that represents a first event type pertinent to a first cluster of items. The machine calculates a probability that a second event type pertaining to a second cluster of items will co-occur with the first event type. The machine identifies an item from the second cluster to be recommended and presents a recommendation of the item to the potential buyer.
摘要翻译：聚类和推荐机器确定项目被包含在项目集群中。机器访问描述项目的项目数据。机器访问表示集群的向量，并根据项目变量和概率参数计算项目包含在集群中的可能性。根据可能性，机器确定该项目包含在群集中。该机器还向潜在买家推荐一个物品。机器访问表示与第一个项目集相关的第一个事件类型的行为数据。机器计算属于第二类项目的第二事件类型与第一事件类型共同出现的概率。机器识别要推荐的第二个集群中的一个项目，并将该项目的建议呈现给潜在的买方。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式