公开(公告)号：US07647314B2

公开(公告)日：2010-01-12

申请号：US11414405

申请日：2006-04-28

申请人： Gordon Sun ,  Zhaohui Zheng

发明人： Gordon Sun ,  Zhaohui Zheng

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/30864

摘要： System and method for the determination of the relevance of a content item to a query through the use of a machine learned relevance function that incorporates click-through features of the content items. A method for selecting a relevance function to determine a relevance of a query-content item pair comprises generating training set having one or more query-URL pairs labeled for relevance based on their click-through features. The labeled query-URL pairs are used to determine the relevance function by minimizing a loss function that accounts for click-through features of the content item. The computed relevance function is then applied to the click-through features of unlabeled content items to assign relevance scores thereto. An inverted click-through index of query-score pairs is formed and combined with the content index to improve relevance of search results.

摘要翻译： 用于通过使用包含内容项目的点击特征的机器学习相关性功能来确定内容项目与查询的相关性的系统和方法。 用于选择相关函数以确定查询内容项目对的相关性的方法包括基于其点击特征生成具有标记为相关性的一个或多个查询 - URL对的训练集合。 标记的查询URL对用于通过最小化考虑内容项目的点击特征的损失函数来确定相关性功能。 然后将所计算的相关性函数应用于未标记的内容项的点击特征以向其分配相关性分数。 形成查询分数对的反转点击索引，并与内容索引相结合，提高搜索结果的相关性。

公开(公告)号：US20090248668A1

公开(公告)日：2009-10-01

申请号：US12060195

申请日：2008-03-31

申请人： Zhaohui Zheng ,  Hongyuan Zha ,  Gordon Sun

发明人： Zhaohui Zheng ,  Hongyuan Zha ,  Gordon Sun

IPC分类号： G06F17/30

CPC分类号： G06F17/30675

摘要： Embodiments of the present invention provide for methods, systems and computer program products for learning ranking functions to determine the ranking of one or more content items that are responsive to a query. The present invention includes generating one or more training sets comprising one or more content item-query pairs and determining one or more contradicting pairs in a given training sets. An optimization function to minimize the number of contradicting pairs in the training set is formulated. and modified by incorporating a grade difference between one or more content items corresponding to the query in the training set and applied to each query in the training set. A ranking function is determined based on the application of regression trees on the queries of the training set minimized by the optimization function and stored for application to content item-query pairs not contained in the one or more training sets.

摘要翻译： 本发明的实施例提供了用于学习排名功能以确定响应于查询的一个或多个内容项目的排名的方法，系统和计算机程序产品。 本发明包括生成包括一个或多个内容项查询对并且确定给定训练集中的一个或多个矛盾对的一个或多个训练集。 制定了最小化训练集中矛盾对数的优化函数。 并且通过在训练集中并入与训练集中的查询相对应的一个或多个内容项之间并且应用于训练集中的每个查询来修改。 基于对优化函数最小化的训练集的查询的回归树的应用来确定排序函数，并将其存储用于不包含在一个或多个训练集中的内容项查询对。

公开(公告)号：US20080208836A1

公开(公告)日：2008-08-28

申请号：US11710097

申请日：2007-02-23

申请人： Zhaohui Zheng ,  Hongyuan Zha ,  Keke Chen ,  Gordon Sun

发明人： Zhaohui Zheng ,  Hongyuan Zha ,  Keke Chen ,  Gordon Sun

IPC分类号： G06F17/30

CPC分类号： G06F16/951

摘要： A method and apparatus for determining a ranking function by regression using relative preference data. A number of iterations are performed in which to following is performed. The current ranking function is used to compare pairs of elements. The comparisons are checked against actual preference data to determine for which pairs the ranking function mis-predicted (contradicting pairs). A regression function is fitted to a set of training data that is based on contradicting pairs and a target value for each element. The target value for each element may be based on the value that the ranking function predicted for the other element in the pair. The ranking function for the next iteration is determined based, at least in part, on the regression function. The final ranking function is established based on the regression functions. For example, the final ranking function may be based on a linear combination of regression functions.

摘要翻译： 一种用于使用相对偏好数据通过回归确定排序函数的方法和装置。 执行多个迭代，其中执行跟随。 当前的排序函数用于比较元素对。 根据实际偏好数据来检查比较，以确定排序功能对于哪些对误预测（矛盾对）。 回归函数适用于基于矛盾对的训练数据集合和每个元素的目标值。 每个元素的目标值可以基于排序函数为该对中的另一元素预测的值。 至少部分地基于回归函数确定下一次迭代的排序函数。 基于回归函数建立最终排名函数。 例如，最终排序函数可以基于回归函数的线性组合。

公开(公告)号：US10685065B2

公开(公告)日：2020-06-16

申请号：US14385274

申请日：2012-03-17

申请人： Zhaohui Zheng ,  Xin Li ,  Rongqing Lu ,  Shuanghong Yang

发明人： Zhaohui Zheng ,  Xin Li ,  Rongqing Lu ,  Shuanghong Yang

IPC分类号： G06F16/9032 ,  G06F16/9535 ,  G06Q30/02 ,  G06N5/04

摘要： Method, system, and programs for recommending content to a user. First information related to one or more previous users is first obtained. A model that maps from users to topics of interest is then established based on the first information related to the one or more previous users. Second information related to the current user is also obtained. One or more topics of interest are identified for the current user based on the model. Content is recommended to the current user in accordance with the one or more topics of interest for the current user. Eventually, an updated model is generated by integrating information associated with the current user with the model established based on the first information related to the one or more previous users. The information associated with the current user includes the second information related to the current user.

公开(公告)号：US10019518B2

公开(公告)日：2018-07-10

申请号：US12577045

申请日：2009-10-09

申请人： Jiang Chen ,  Wei Chu ,  Zhenzhen Kou ,  Zhaohui Zheng

发明人： Jiang Chen ,  Wei Chu ,  Zhenzhen Kou ,  Zhaohui Zheng

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F16/951

摘要： Methods and systems are disclosed that relate to ranking functions for multiple different domains. By way of example but not limitation, ranking functions for multiple different domains may be trained based on inter-domain loss, and such ranking functions may be used to rank search results from multiple different domains so that they may be blended without normalizing relevancy scores.

公开(公告)号：US09536003B2

公开(公告)日：2017-01-03

申请号：US14385220

申请日：2012-03-17

申请人： Zhaohui Zheng ,  Rongqing Lu ,  Xin Li

发明人： Zhaohui Zheng ,  Rongqing Lu ,  Xin Li

IPC分类号： G06F17/30 ,  G06F7/00 ,  G06F21/62

CPC分类号： G06F17/30696 ,  G06F17/30011 ,  G06F17/30312 ,  G06F17/30619 ,  G06F17/30702 ,  G06F17/30705 ,  G06F17/30864 ,  G06F17/30867 ,  G06F21/6245

摘要： Method, system, and programs for hybrid information query. A request is first received from a user associated with a hybrid query. The hybrid query is expressed in accordance with an input in terms of one of a user, a feature, and a document, and a desired hybrid query result in terms of one of a user, a feature, and a document. A mapping is then determined between the input and the desired hybrid query result. A hybrid model is established based on hybrid information collected and associated with one or more users. The mapping is performed based on the hybrid model to obtain the desired hybrid query result based on the input. Eventually, the desired hybrid query result is provided as a response to the hybrid query.

摘要翻译： 混合信息查询的方法，系统和程序。 首先从与混合查询相关联的用户接收到请求。 根据用户，特征和文档之一的输入以及用户，特征和文档之一的期望的混合查询结果来表示混合查询。 然后在输入和期望的混合查询结果之间确定映射。 基于收集并与一个或多个用户相关联的混合信息建立混合模型。 基于混合模型进行映射，以基于输入获得所需的混合查询结果。 最终，期望的混合查询结果作为对混合查询的响应被提供。

公开(公告)号：US08583502B2

公开(公告)日：2013-11-12

申请号：US12838169

申请日：2010-07-16

申请人： Narayanan Sadagopan ,  Zhaohui Zheng

发明人： Narayanan Sadagopan ,  Zhaohui Zheng

IPC分类号： G06Q30/00

CPC分类号： G06F17/30867 ,  G06F17/30657 ,  G06F17/30684 ,  G06Q30/02 ,  G06Q30/0641

摘要： A server determines a plurality of immediate candidate items for a first web page to recommend to a user. For each particular immediate candidate item of the plurality of immediate candidate items, the server determines a separate sequence of two or more subsequent possible candidate items for subsequent web pages to recommend to the user in the event that the user selects the particular immediate candidate item. Further, the server selects a particular immediate candidate item from the plurality of immediate candidate items for the first web page to recommend to the user. The first web page that recommends the plurality of immediate candidate items is generated and sent over the Internet to the user.

摘要翻译： 服务器确定用于第一网页的多个即时候选项目以推荐给用户。 对于多个即时候选项目中的每个特定即时候选项目，服务器确定用于后续网页的两个或更多个后续可能候选项目的单独序列，以在用户选择特定直接候选项目的情况下推荐给用户。 此外，服务器从第一网页的多个直接候选项目中选择特定的即时候选项目以向用户推荐。 建立多个即时候选项目的第一网页通过因特网生成并发送给用户。

公开(公告)号：US20130297688A1

公开(公告)日：2013-11-07

申请号：US13463109

申请日：2012-05-03

申请人： Zhaohui Zheng

发明人： Zhaohui Zheng

IPC分类号： G06F15/16

CPC分类号： G06Q50/01 ,  G06Q30/0205 ,  G06Q30/0282 ,  G06Q30/0631

摘要： Performing an on-line recommendation includes: analyzing real-time data from various sources; determining, from the analysis, events in which a user may be interested; extracting the determined events; storing the extracted events in a data store; and performing a recommendation function. The recommendation function includes: ranking the extracted events to determine the events in which the user is most likely to be interested; and performing location-based filtering, retaining those extracted events that are within a geo-location range proximate to the user, thus generating optimal events.

摘要翻译： 执行在线建议包括：分析来自各种来源的实时数据; 从分析中确定用户可能感兴趣的事件; 提取确定的事件; 将所提取的事件存储在数据存储器中; 并执行推荐功能。 推荐功能包括：对提取的事件进行排名，以确定用户最有可能感兴趣的事件; 并且执行基于位置的过滤，保留在靠近用户的地理位置范围内的提取的事件，从而产生最佳事件。

公开(公告)号：US08429027B2

公开(公告)日：2013-04-23

申请号：US12941842

申请日：2010-11-08

申请人： Zhaohui Zheng

发明人： Zhaohui Zheng

IPC分类号： G06Q30/00

CPC分类号： G06Q50/01 ,  G06Q30/02

摘要： Particular embodiments extract a plurality of users, a plurality of establishments, and a plurality of items from dining information provided by at least one of the plurality of users, each of the plurality of establishments sells food or beverage; construct a user-establishment matrix, a user-item matrix, and an establishment-item matrix using the plurality of users, the plurality of establishments, and the plurality of items; generate a user latent representation for the plurality of users, an establishment latent representation for the plurality of establishments, and an item latent representation for the plurality of items; and compute one or more correlations using the user latent representation, the establishment latent representation, or the item latent representation, wherein each of the one or more correlations is between two users, two establishments, two items, one user and one establishment, one of user and one item, or one establishment and one item.

摘要翻译： 具体实施例从多个用户中的至少一个提供的餐饮信息中提取多个用​​户，多个机构和多个项目，多个机构中的每一个销售食品或饮料; 使用所述多个用户，所述多个机构和所述多个项目来构建用户建立矩阵，用户项目矩阵和建立项目矩阵; 为所述多个用户生成用户潜在表示，所述多个机构的企业潜在表示以及所述多个项目的项目潜在表示; 并且使用所述用户潜在表示，所述建立潜在表示或所述项目潜在表示来计算一个或多个相关性，其中所述一个或多个相关性中的每一个在两个用户，两个机构，两个项目，一个用户和一个机构之间， 用户和一个项目，或一个企业和一个项目。

公开(公告)号：US07685078B2

公开(公告)日：2010-03-23

申请号：US11755134

申请日：2007-05-30

申请人： Keke Chen ,  Ya Zhang ,  Zhaohui Zheng ,  Hongyuan Zha ,  Gordon Sun

发明人： Keke Chen ,  Ya Zhang ,  Zhaohui Zheng ,  Hongyuan Zha ,  Gordon Sun

IPC分类号： G06N5/00

CPC分类号： G06F17/30864

摘要： The present invention relates to systems and methods for determining a content item relevance function. The method comprises collecting user preference data at a search provider for storage in a user preference data store and collecting expert-judgment data at the search provider for storage in an expert sample data store. A modeling module trains a base model through the use of the expert-judgment data and tunes the base model through the use of the user preference data to learn a set of one or more tuned models. A measure (B measure) is designed to evaluate the balanced performance of tuned model over expert judgment and user preference. The modeling module generates or selects the content item relevance function from the tuned models with B measure as the selection criterion.

摘要翻译： 本发明涉及用于确定内容项相关性功能的系统和方法。 该方法包括在搜索提供者处收集用户偏好数据以存储在用户偏好数据存储中，并在搜索提供商处收集专家判断数据以存储在专家样本数据存储中。 建模模块通过使用专家判断数据来训练基本模型，并通过使用用户偏好数据来调整基本模型，以学习一组或多个调谐模型。 测量（B测量）旨在评估调谐模型与专家判断和用户偏好的平衡性能。 建模模块从具有B测量的调谐模型生成或选择内容项相关性函数作为选择标准。