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. 发明授权

US08661042B2 Collaborative filtering with hashing 有权
标题翻译：用哈希进行协同过滤
公开(公告)号：US08661042B2
公开(公告)日：2014-02-25
申请号：US12906551
申请日：2010-10-18
申请人： Martin B. Scholz , Shyamsundar Rajaram , Rajan Lukose
发明人： Martin B. Scholz , Shyamsundar Rajaram , Rajan Lukose
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/30699
摘要： Systems, methods, and machine readable and executable instructions are provided for collaborative filtering. Collaborative filtering includes representing users and objects by rows and columns in a binary ratings matrix having a particular dimensional space. Unknown values in the binary ratings matrix are weighted with a weight matrix having the particular dimensional space. The binary ratings matrix and the weight matrix are hashed into a lower dimensional space by one of row and column. The hashed binary ratings matrix and the hashed weight matrix are low-rank approximated by alternating least squares. A result of the low-rank approximation for the one of row and column is updated using the binary ratings matrix and the weight matrix. A recommendation of one of the objects can be generated for one of the users based on the updated result.
摘要翻译：提供系统，方法和机器可读和可执行指令用于协同过滤。协同过滤包括用具有特定尺寸空间的二进制评级矩阵中的行和列来表示用户和对象。二进制等级矩阵中的未知值用具有特定尺寸空间的权重矩阵加权。二进制等级矩阵和权重矩阵通过行和列之一被散列成较低维的空间。散列二进制等级矩阵和散列权重矩阵是通过交替的最小二乘法近似的低阶。使用二进制等级矩阵和权重矩阵来更新行和列之一的低阶近似的结果。可以基于更新的结果为一个用户生成其中一个对象的推荐。

2. 发明申请

US20120096009A1 COLLABORATIVE FILTERING WITH HASHING 有权
标题翻译：与洗涤的协同过滤
公开(公告)号：US20120096009A1
公开(公告)日：2012-04-19
申请号：US12906551
申请日：2010-10-18
申请人： Martin B. Scholz , Shyamsundar Rajaram , Rajant Lukose
发明人： Martin B. Scholz , Shyamsundar Rajaram , Rajant Lukose
IPC分类号： G06F17/30
CPC分类号： G06F17/30699
摘要： Systems, methods, and machine readable and executable instructions are provided for collaborative filtering. Collaborative filtering includes representing users and objects by rows and columns in a binary ratings matrix having a particular dimensional space. Unknown values in the binary ratings matrix are weighted with a weight matrix having the particular dimensional space. The binary ratings matrix and the weight matrix are hashed into a lower dimensional space by one of row and column. The hashed binary ratings matrix and the hashed weight matrix are low-rank approximated by alternating least squares. A result of the low-rank approximation for the one of row and column is updated using the binary ratings matrix and the weight matrix. A recommendation of one of the objects can be generated for one of the users based on the updated result.
摘要翻译：提供系统，方法和机器可读和可执行指令用于协同过滤。协同过滤包括用具有特定尺寸空间的二进制评级矩阵中的行和列来表示用户和对象。二进制等级矩阵中的未知值用具有特定尺寸空间的权重矩阵加权。二进制等级矩阵和权重矩阵通过行和列之一被散列成较低维的空间。散列二进制等级矩阵和散列权重矩阵是通过交替的最小二乘法近似的低阶。使用二进制等级矩阵和权重矩阵来更新行和列之一的低阶近似的结果。可以基于更新的结果为一个用户生成其中一个对象的推荐。

3. 发明申请

US20100082579A1 Identification Of Data Objects Within A Computer Database 有权
标题翻译：计算机数据库中数据对象的识别
公开(公告)号：US20100082579A1
公开(公告)日：2010-04-01
申请号：US12243075
申请日：2008-10-01
申请人： Shyamsundar Rajaram
发明人： Shyamsundar Rajaram
IPC分类号： G06F17/30
CPC分类号： G06F17/3033 , G06F17/30528 , G06F17/30598 , G06F17/30607 , G06F17/30864
摘要： Provided are, among other things, systems, methods and techniques for identifying matching objects in a computer database. In one representative technique, a set of attribute-value pairs corresponding to a query data object are input, with individual ones of the attribute-value pairs including an identified attribute and a value for the identified attribute; multiple characteristic fingerprints are assigned to individual ones of the attribute-value pairs in the set, the characteristic fingerprints having been selected from an attribute-specific field of available characteristic fingerprints based on the value for the identified attribute; a subset of at least one characteristic fingerprint is selected from across the characteristic fingerprints for the query data object, based on a selection criterion, and a database is queried using the subset of at least one characteristic fingerprint to identify any matches.
摘要翻译：除其他之外，提供用于识别计算机数据库中的匹配对象的系统，方法和技术。在一种代表性技术中，输入与查询数据对象对应的一组属性值对，其中属性值对中的各个属性值对包括识别的属性和所识别的属性的值; 将多个特征指纹分配给集合中的属性值对的各个特征指纹，基于所识别的属性的值，已经从可用特征指纹的属性特定字段中选择了特征指纹; 基于选择标准，从查询数据对象的特征指纹中选择至少一个特征指纹的子集，并且使用至少一个特征指纹的子集来查询数据库以识别任何匹配。

4. 发明授权

US09141687B2 Identification of data objects within a computer database 有权
标题翻译：识别计算机数据库中的数据对象
公开(公告)号：US09141687B2
公开(公告)日：2015-09-22
申请号：US12243075
申请日：2008-10-01
申请人： Shyamsundar Rajaram
发明人： Shyamsundar Rajaram
IPC分类号： G06F17/30
CPC分类号： G06F17/3033 , G06F17/30528 , G06F17/30598 , G06F17/30607 , G06F17/30864
摘要： Provided are, among other things, systems, methods and techniques for identifying matching objects in a computer database. In one representative technique, a set of attribute-value pairs corresponding to a query data object are input, with individual ones of the attribute-value pairs including an identified attribute and a value for the identified attribute; multiple characteristic fingerprints are assigned to individual ones of the attribute-value pairs in the set, the characteristic fingerprints having been selected from an attribute-specific field of available characteristic fingerprints based on the value for the identified attribute; a subset of at least one characteristic fingerprint is selected from across the characteristic fingerprints for the query data object, based on a selection criterion, and a database is queried using the subset of at least one characteristic fingerprint to identify any matches.
摘要翻译：除其他之外，提供用于识别计算机数据库中的匹配对象的系统，方法和技术。在一种代表性技术中，输入与查询数据对象对应的一组属性值对，其中属性值对中的各个属性值对包括识别的属性和所识别的属性的值; 将多个特征指纹分配给集合中的属性值对的各个特征指纹，基于所识别的属性的值，已经从可用特征指纹的属性特定字段中选择了特征指纹; 基于选择标准，从查询数据对象的特征指纹中选择至少一个特征指纹的子集，并且使用至少一个特征指纹的子集来查询数据库以识别任何匹配。

5. 发明申请

US20110113385A1 VISUALLY REPRESENTING A HIERARCHY OF CATEGORY NODES 有权
标题翻译：可视地表示类别编号
公开(公告)号：US20110113385A1
公开(公告)日：2011-05-12
申请号：US12614193
申请日：2009-11-06
申请人： Craig Peter Sayers , Ignacio Zendejas , Rajan Lukose , Martin Scholz , Shyamsundar Rajaram
发明人： Craig Peter Sayers , Ignacio Zendejas , Rajan Lukose , Martin Scholz , Shyamsundar Rajaram
IPC分类号： G06F3/048
CPC分类号： G06F17/30873
摘要： In a method of visually representing a hierarchy of category nodes that identify one or more concepts, accesses to documents are tracked and one or more concepts that are relevant to the accessed documents are determined. In addition, one or more category paths through the hierarchy of category nodes for the determined one or more concepts are determined and relevance levels for the category nodes with respect to the determined one or more concepts are determined based upon the generated one or more category paths. Moreover, a graph depicting a visual representation of the relevance levels determined for each of the category nodes in the hierarchy of category nodes is constructed.
摘要翻译：在可视地表示标识一个或多个概念的类别节点的层次结构的方法中，跟踪对文档的访问，并确定与被访问文档相关的一个或多个概念。此外，确定用于确定的一个或多个概念的通过类别节点的层次结构的一个或多个类别路径，并且基于所生成的一个或多个类别路径来确定关于所确定的一个或多个概念的类别节点的相关性级别。此外，构造描绘对于类别节点的层次结构中的每个类别节点确定的相关性级别的视觉表示的图形。

6. 发明申请

US20110112824A1 DETERMINING AT LEAST ONE CATEGORY PATH FOR IDENTIFYING INPUT TEXT 审中-公开
标题翻译：确定至少一个类别路径识别输入文本
公开(公告)号：US20110112824A1
公开(公告)日：2011-05-12
申请号：US12614260
申请日：2009-11-06
申请人： Craig Peter Sayers , Ignacio Zendejas , Rajan Lukose , Martin Scholz , Shyamsundar Rajaram
发明人： Craig Peter Sayers , Ignacio Zendejas , Rajan Lukose , Martin Scholz , Shyamsundar Rajaram
IPC分类号： G06F17/27
CPC分类号： G06F17/27
摘要： In a method of determining at least one category path for identifying an input text, one or more categories that are most relevant to the input text are determined, one or more concepts that are most relevant to the input text using information from a labeled text data source and the one or more categories determined to be the most relevant to the input text are determined, and one or more category paths through a hierarchy of predefined category levels are determined for one or more of the determined concepts.
摘要翻译：在确定用于识别输入文本的至少一个类别路径的方法中，确定与输入文本最相关的一个或多个类别，使用来自标记文本数据的信息与输入文本最相关的一个或多个概念源和确定为与输入文本最相关的一个或多个类别，并且针对一个或多个所确定的概念确定通过预定类别级别的层级的一个或多个类别路径。

7. 发明授权

US08954893B2 Visually representing a hierarchy of category nodes 有权
标题翻译：目视表示类别节点的层次结构
公开(公告)号：US08954893B2
公开(公告)日：2015-02-10
申请号：US12614193
申请日：2009-11-06
申请人： Craig Peter Sayers , Ignacio Zendejas , Rajan Lukose , Martin Scholz , Shyamsundar Rajaram
发明人： Craig Peter Sayers , Ignacio Zendejas , Rajan Lukose , Martin Scholz , Shyamsundar Rajaram
IPC分类号： G06F3/048 , G06F17/30 , G06F3/00
CPC分类号： G06F17/30873
摘要： In a method of visually representing a hierarchy of category nodes that identify one or more concepts, accesses to documents are tracked and one or more concepts that are relevant to the accessed documents are determined. In addition, one or more category paths through the hierarchy of category nodes for the determined one or more concepts are determined and relevance levels for the category nodes with respect to the determined one or more concepts are determined based upon the generated one or more category paths. Moreover, a graph depicting a visual representation of the relevance levels determined for each of the category nodes in the hierarchy of category nodes is constructed.
摘要翻译：在可视地表示标识一个或多个概念的类别节点的层次结构的方法中，跟踪对文档的访问，并确定与被访问文档相关的一个或多个概念。此外，确定用于确定的一个或多个概念的通过类别节点的层次结构的一个或多个类别路径，并且基于所生成的一个或多个类别路径来确定关于所确定的一个或多个概念的类别节点的相关性级别。此外，构造描绘对于类别节点的层次结构中的每个类别节点确定的相关性级别的视觉表示的图形。

8. 发明授权

US08631017B2 Collaborative filtering with hashing 有权
标题翻译：用哈希进行协同过滤
公开(公告)号：US08631017B2
公开(公告)日：2014-01-14
申请号：US12970262
申请日：2010-12-16
申请人： Martin B. Scholz , Shyamsundar Rajaram , Rajan Lukose
发明人： Martin B. Scholz , Shyamsundar Rajaram , Rajan Lukose
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/3053
摘要： Systems, methods, and machine readable and executable instructions are provided for collaborative filtering. Collaborative filtering includes representing users and objects by rows and columns in an ordinal ratings matrix having a particular dimensional space. Values in the ordinal ratings matrix are weighted with a weight matrix having the particular dimensional space. The weight matrix is hashed into a lower dimensional space by one of row and column by multiplying a projection matrix by the weight matrix. The ordinal ratings matrix is hashed into a lower dimensional space by multiplying the projection matrix by an element-wise product of the weight matrix and the ordinal ratings matrix to form a reduced ratings matrix, and element-wise dividing the reduced ratings matrix by the hashed weight matrix. The hashed ordinal ratings matrix and the hashed weight matrix are low-rank approximated by alternating least squares. A result of the low-rank approximation for the one of row and column is updated using the ordinal ratings matrix and the weight matrix. A recommendation of one of the objects can be generated for one of the users based on the updated result.
摘要翻译：提供系统，方法和机器可读和可执行指令用于协同过滤。协作过滤包括用具有特定尺寸空间的序数等级矩阵中的行和列表示用户和对象。使用具有特定尺寸空间的权重矩阵对序数等级矩阵中的值进行加权。通过将投影矩阵乘以权重矩阵，权重矩阵通过行和列之一被散列成较低维空间。通过将投影矩阵乘以权重矩阵和序数等级矩阵的元素乘积来将序数等级矩阵散列到较低维空间中，以形成减小的等级矩阵，并且将分数矩阵除以散列重量矩阵。散列序数等级矩阵和散列权重矩阵由交替的最小二乘法近似近似。使用序数等级矩阵和权重矩阵来更新行和列之一的低阶近似的结果。可以基于更新的结果为一个用户生成其中一个对象的推荐。

9. 发明申请

US20060164997A1 Dependency structure from temporal data 有权
标题翻译：时态数据依赖结构
公开(公告)号：US20060164997A1
公开(公告)日：2006-07-27
申请号：US11027188
申请日：2004-12-30
申请人： Thore Graepel , Ralf Herbrich , Shyamsundar Rajaram
发明人： Thore Graepel , Ralf Herbrich , Shyamsundar Rajaram
IPC分类号： H04L12/26 , H04J1/16 , H04J3/14 , H04L1/00 , G01R31/08 , G06F11/00 , G08C15/00
CPC分类号： G06F11/3452 , G06F11/3419 , G06F11/3495 , G06F2201/835
摘要： Based on the time series data from multiple components, the systems administrator or other managing entity may desire to find the temporal dependencies between the different time series data over time. For example, based on actions indicated in time series data from two or more servers in a server network, a dependency structure may be determined which indicates a parent/child or dependent relationship between the two or more servers. In some cases, it may also be beneficial to predict the state of a child component, and/or predict the average time to a state change or event of a child component based on the parent time series data. These determinations and predications may reflect the logical connections between actions of components. The relationships and/or predictions may be expressed graphically and/or in terms of a probability distribution.
摘要翻译：基于来自多个组件的时间序列数据，系统管理员或其他管理实体可能希望随着时间的推移找到不同时间序列数据之间的时间依赖性。例如，基于来自服务器网络中的两个或多个服务器的时间序列数据中指示的动作，可以确定依赖结构，其指示两个或多个服务器之间的父/子或依赖关系。在某些情况下，基于父时间序列数据预测子组件的状态和/或预测子组件的状态改变或事件的平均时间也可能是有益的。这些确定和预测可能反映组件的动作之间的逻辑连接。关系和/或预测可以以图形和/或以概率分布来表示。

10. 发明申请

US20050228673A1 Techniques for separating and evaluating audio and video source data 审中-公开
标题翻译：分离和评估音频和视频源数据的技术
公开(公告)号：US20050228673A1
公开(公告)日：2005-10-13
申请号：US10813642
申请日：2004-03-30
申请人： Ara Nefian , Shyamsundar Rajaram
发明人： Ara Nefian , Shyamsundar Rajaram
IPC分类号： G06K9/00 , G10L15/24 , G10L21/00
CPC分类号： G10L15/25
摘要： Methods, systems, and apparatus are provided to separate and evaluate audio and video. Audio and video are captured; the audio is evaluated to detect one or more speakers speaking. Visual features are associated with the speakers speaking. The audio and video are separated and corresponding portions of the audio are mapped to the visual features for purposes of isolating audio associated with each speaker and for purposes of filtering out noise associated with the audio.
摘要翻译：提供方法，系统和设备来分离和评估音频和视频。音频和视频被捕获; 评估音频以检测一个或多个说话者。可视功能与扬声器相关联。音频和视频被分离，音频的相应部分被映射到视觉特征，用于隔离与每个扬声器相关联的音频，并且用于滤除与音频相关联的噪声。

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