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

US20120197834A1 ESTIMATING RELATEDNESS IN SOCIAL NETWORK 有权
标题翻译：评估社会网络中的相关性
公开(公告)号：US20120197834A1
公开(公告)日：2012-08-02
申请号：US13018424
申请日：2011-02-01
申请人： Rina Panigrahy , Mikhail Kapralov
发明人： Rina Panigrahy , Mikhail Kapralov
IPC分类号： G06N5/02
CPC分类号： G06N5/048 , G06F17/30958 , G06Q50/01
摘要： To facilitate the estimation of relatedness between nodes of a graph, implementations estimate relatedness between nodes in a graph by pre-computing for a subset of sample nodes (e.g., center nodes) a plurality of transition probabilities between each sample node and each of the other nodes in the graph, and then later when queried the implementations calculate in real-time the resultant estimated transition probability between the first node and the second node through the at least one sample node based on the pre-computed transition probabilities.
摘要翻译：为了便于对图的节点之间的相关性的估计，实现通过对样本节点（例如，中心节点）的子集进行预先计算来估计图中的节点之间的相关性，每个样本节点与其他每个样本节点之间的多个转换概率然后在查询时，实施方式基于预先计算的转移概率，通过至少一个采样节点实时计算第一节点和第二节点之间的估计转移概率。

2. 发明授权

US08862528B2 Predicting values in sequence 有权
标题翻译：按顺序预测值
公开(公告)号：US08862528B2
公开(公告)日：2014-10-14
申请号：US13105908
申请日：2011-05-12
申请人： Rina Panigrahy , Mikhail Kapralov
发明人： Rina Panigrahy , Mikhail Kapralov
IPC分类号： G06N5/00 , H04L27/06 , G06N99/00
CPC分类号： H04L27/06 , G06N99/00
摘要： Multiple data prediction strategies are received. Each data prediction strategy may predict a next data value in a sequence of data values with a corresponding confidence value. Rather than rely on a single prediction strategy, the predictions of each of the data prediction strategies are linearly combined to generate a single prediction that is more accurate and has a lower overall loss than any of the individual prediction strategies. Further, a deviation is calculated based on the values in the sequence of values that have been observed so far using a weighted sum that favors more recent values in the sequence over less recent values in the sequence. A prediction of the next value in the sequence is generated based on the combined strategies and the calculated deviation.
摘要翻译：收到多个数据预测策略。每个数据预测策略可以用对应的置信度值来预测数据值序列中的下一个数据值。不依赖于单一预测策略，每个数据预测策略的预测是线性组合的，以产生比任何单个预测策略更准确且总体损失更低的单个预测。此外，基于迄今为止已经观察到的值序列中的值来计算偏差，该加权和有利于序列中较近的值中的更新的近似值。基于组合策略和计算出的偏差，生成序列中下一个值的预测。

3. 发明授权

US08719211B2 Estimating relatedness in social network 有权
标题翻译：估计社会网络中的相关性
公开(公告)号：US08719211B2
公开(公告)日：2014-05-06
申请号：US13018424
申请日：2011-02-01
申请人： Rina Panigrahy , Mikhail Kapralov
发明人： Rina Panigrahy , Mikhail Kapralov
IPC分类号： G06F9/44 , G06F15/18 , G06N5/04
CPC分类号： G06N5/048 , G06F17/30958 , G06Q50/01
摘要： To facilitate the estimation of relatedness between nodes of a graph, implementations estimate relatedness between nodes in a graph by pre-computing for a subset of sample nodes (e.g., center nodes) a plurality of transition probabilities between each sample node and each of the other nodes in the graph, and then later when queried the implementations calculate in real-time the resultant estimated transition probability between the first node and the second node through the at least one sample node based on the pre-computed transition probabilities.
摘要翻译：为了便于对图的节点之间的相关性的估计，实现通过对样本节点（例如，中心节点）的子集进行预先计算来估计图中的节点之间的相关性，每个样本节点与其他每个样本节点之间的多个转换概率然后在查询时，实施方式基于预先计算的转移概率，通过至少一个采样节点实时计算第一节点和第二节点之间的估计转移概率。

4. 发明申请

US20120288036A1 PREDICTING VALUES IN SEQUENCE 有权
标题翻译：序列中的预测值
公开(公告)号：US20120288036A1
公开(公告)日：2012-11-15
申请号：US13105908
申请日：2011-05-12
申请人： Rina Panigrahy , Mikhail Kapralov
发明人： Rina Panigrahy , Mikhail Kapralov
IPC分类号： H04L27/06
CPC分类号： H04L27/06 , G06N99/00
摘要： Multiple data prediction strategies are received. Each data prediction strategy may predict a next data value in a sequence of data values with a corresponding confidence value. Rather than rely on a single prediction strategy, the predictions of each of the data prediction strategies are linearly combined to generate a single prediction that is more accurate and has a lower overall loss than any of the individual prediction strategies. Further, a deviation is calculated based on the values in the sequence of values that have been observed so far using a weighted sum that favors more recent values in the sequence over less recent values in the sequence. A prediction of the next value in the sequence is generated based on the combined strategies and the calculated deviation.
摘要翻译：收到多个数据预测策略。每个数据预测策略可以用对应的置信度值来预测数据值序列中的下一个数据值。不依赖单一预测策略，每个数据预测策略的预测都是线性组合的，以产生比任何单个预测策略更准确且总体损失更低的单个预测。此外，基于迄今为止已经观察到的值序列中的值来计算偏差，该加权和有利于序列中较近的值中的更新的近似值。基于组合策略和计算出的偏差，生成序列中下一个值的预测。

5. 发明申请

US20120299925A1 DETERMINING AFFINITY IN SOCIAL NETWORKS 审中-公开
标题翻译：确定社会网络的重要性
公开(公告)号：US20120299925A1
公开(公告)日：2012-11-29
申请号：US13113103
申请日：2011-05-23
申请人： Marc A. Najork , Rina Panigrahy
发明人： Marc A. Najork , Rina Panigrahy
IPC分类号： G06T11/20 , G06F15/16
CPC分类号： G06Q50/01 , G06T11/206
摘要： A graph is generated based on a social networking application with a node for each user account, and one or more edges representing the social networking relationships between the user accounts (e.g., friends). A sketch is generated for each node in iterations where edges are removed from the graph and a set of reachable nodes is determined for the node. A representative node is then selected from the set of reachable nodes and added to the sketch as a dimension. The generated sketches for two nodes are used to calculate an affinity score between the accounts associated with each of the two nodes.
摘要翻译：基于具有每个用户帐户的节点的社交网络应用以及表示用户帐户（例如，朋友）之间的社交网络关系的一个或多个边缘生成图。在从图中移除边缘的迭代中为每个节点生成草图，并为该节点确定一组可到达节点。然后从可达节点集合中选择一个代表性节点，并将其作为维度添加到草图中。用于两个节点的生成的草图用于计算与两个节点中的每一个相关联的帐户之间的亲和度分数。

6. 发明授权

US08099417B2 Semi-supervised part-of-speech tagging 有权
标题翻译：半监督的词性标签
公开(公告)号：US08099417B2
公开(公告)日：2012-01-17
申请号：US11954216
申请日：2007-12-12
申请人： Sreenivas Gollapudi , Rina Panigrahy
发明人： Sreenivas Gollapudi , Rina Panigrahy
IPC分类号： G06F17/30
CPC分类号： G06F17/30864
摘要： Relevant search results for a given query may be determined using click data for the query and the number of times the query is issued to a search engine. The number of clicks that a result receives for the given query may provide a feedback mechanism to the search engine on how relevant the result is for the given query. The frequency of a query along with the associated clicks provides the search engine with the effectiveness of the query in producing relevant results. Edges in a graph of queries versus results may be weighted in accordance with the click data and the efficiency to rank the search results provided to a user.
摘要翻译：给定查询的相关搜索结果可以使用查询的点击数据和查询发布到搜索引擎的次数来确定。结果为给定查询获得的点击次数可以向搜索引擎提供关于给定查询的结果的相关性的反馈机制。查询的频率以及相关的点击次数为搜索引擎提供了生成相关结果时查询的有效性。可以根据点击数据和提供给用户的搜索结果的效率来加权查询与结果图表中的边。

7. 发明申请

US20100281478A1 MULTIPHASE VIRTUAL MACHINE HOST CAPACITY PLANNING 审中-公开
标题翻译：多功能虚拟机主机容量规划
公开(公告)号：US20100281478A1
公开(公告)日：2010-11-04
申请号：US12433919
申请日：2009-05-01
申请人： Larry Jay Sauls , Sanjay Gautam , Ehud Wieder , Rina Panigrahy , Kunal Talwar
发明人： Larry Jay Sauls , Sanjay Gautam , Ehud Wieder , Rina Panigrahy , Kunal Talwar
IPC分类号： G06F9/455
CPC分类号： G06F9/5077
摘要： A virtual machine distribution system is described herein that uses a multiphase approach that provides a fast layout of virtual machines on physical computers followed by at least one verification phase that verifies that the layout is correct. During the fast layout phase, the system uses a dimension-aware vector bin-packing algorithm to determine an initial fit of virtual machines to physical hardware based on rescaled resource utilizations calculated against hardware models. During the verification phase, the system uses a virtualization model to check the recommended fit of virtual machine guests to physical hosts created during the fast layout phase to ensure that the distribution will not over-utilize any host given the overhead associated with virtualization. The system modifies the layout to eliminate any identified overutilization. Thus, the virtual machine distribution system provides the advantages of a fast, automated layout planning process with the robustness of slower, exhaustive processes.
摘要翻译：本文描述了一种虚拟机分配系统，其使用多相方法，其在物理计算机上提供虚拟机的快速布局，之后是验证布局正确的至少一个验证阶段。在快速布局阶段，系统使用维度感知向量二进制包装算法，根据硬件模型计算的重新定义的资源利用率来确定虚拟机对物理硬件的初始拟合。在验证阶段，系统使用虚拟化模型来检查虚拟机guest虚拟机对于在快速布局阶段期间创建的物理主机的推荐配置，以确保分配不会过度利用任何与虚拟化相关的开销的主机。系统修改布局以消除任何确定的过度利用。因此，虚拟机分配系统提供了快速，自动化的布局规划过程的优点，具有较慢，详尽的过程的鲁棒性。

8. 发明申请

US20090265786A1 AUTOMATIC BOTNET SPAM SIGNATURE GENERATION 审中-公开
标题翻译：自动BOTNET垃圾邮件签名生成
公开(公告)号：US20090265786A1
公开(公告)日：2009-10-22
申请号：US12104441
申请日：2008-04-17
申请人： Yinglian Xie , Fang Yu , Kannan Achan , Rina Panigrahy , Ivan Osipkov , Geoffrey J. Hulten
发明人： Yinglian Xie , Fang Yu , Kannan Achan , Rina Panigrahy , Ivan Osipkov , Geoffrey J. Hulten
IPC分类号： G06F21/00
CPC分类号： H04L63/1441 , G06F21/564 , G06F2221/2145 , H04L51/12 , H04L63/126 , H04L2463/144
摘要： A framework may be used for generating URL signatures to identify botnet spam and membership. The framework may take a set of unlabeled emails as input that are grouped based on URLs contained within the emails. The framework may return a set of spam URL signatures and a list of corresponding botnet host IP addresses by analyzing the URLs within the emails that are contained within the groups. Each URL signature may be in the form of either a complete URL string or a URL regular expression. The signatures may be used to identify spam emails launched from botnets, while the knowledge of botnet host identities can help filter other spam emails also sent by them.
摘要翻译：一个框架可以用于生成URL签名来识别僵尸网络垃圾邮件和会员资格。框架可以采用一组未标记的电子邮件作为基于邮件中包含的URL分组的输入。框架可以通过分析包含在组内的电子邮件中的URL来返回一组垃圾邮件URL签名和相应僵尸网络主机IP地址的列表。每个URL签名可以是完整的URL字符串或URL正则表达式的形式。签名可用于识别从僵尸网络发起的垃圾邮件，而僵尸网络主机身份的知识可以帮助过滤他们发送的其他垃圾邮件。

9. 发明申请

US20090157643A1 SEMI-SUPERVISED PART-OF-SPEECH TAGGING 有权
标题翻译：半监督的部分话题标签
公开(公告)号：US20090157643A1
公开(公告)日：2009-06-18
申请号：US11954216
申请日：2007-12-12
申请人： Sreenivas Gollapudi , Rina Panigrahy
发明人： Sreenivas Gollapudi , Rina Panigrahy
IPC分类号： G06F17/30
CPC分类号： G06F17/30864
摘要： Relevant search results for a given query may be determined using click data for the query and the number of times the query is issued to a search engine. The number of clicks that a result receives for the given query may provide a feedback mechanism to the search engine on how relevant the result is for the given query. The frequency of a query along with the associated clicks provides the search engine with the effectiveness of the query in producing relevant results. Edges in a graph of queries versus results may be weighted in accordance with the click data and the efficiency to rank the search results provided to a user.
摘要翻译：给定查询的相关搜索结果可以使用查询的点击数据和查询发布到搜索引擎的次数来确定。结果为给定查询获得的点击次数可以向搜索引擎提供关于给定查询的结果的相关性的反馈机制。查询的频率以及相关的点击次数为搜索引擎提供了生成相关结果时查询的有效性。可以根据点击数据和提供给用户的搜索结果的效率来加权查询与结果图表中的边。

10. 发明授权

US07299317B1 Assigning prefixes to associative memory classes based on a value of a last bit of each prefix and their use including but not limited to locating a prefix and for maintaining a Patricia tree data structure 有权
标题翻译：基于每个前缀的最后一位的值分配前缀到关联存储器类，并且包括但不限于定位前缀和维护Patricia树数据结构
公开(公告)号：US07299317B1
公开(公告)日：2007-11-20
申请号：US10165572
申请日：2002-06-08
申请人： Rina Panigrahy , Samar Sharma
发明人： Rina Panigrahy , Samar Sharma
IPC分类号： G06F12/00 , G06F13/00 , G06F13/28
CPC分类号： G11C15/00
摘要： Methods and apparatus are disclosed for maintaining and using entries in one or more associative memories. A last bit of a prefix is checked, and based on this result, the entry is placed into one of two classes of associative memory entries. The entry can then be identified within the associative memory by performing a lookup operation using the prefix padded with ones if the last bit of the prefix is zero or with zeros if the last bit is one to fill out the remaining bits of the lookup word. Entries of different classes of entries are typically either stored in different associative memories, or in the same associative memory with each of these entries including a class specific identifier. Among other applications, these entries can be used to locate a prefix and to store a tree data structure in the one or more associative memories.
摘要翻译：公开了用于维护和使用一个或多个关联存储器中的条目的方法和装置。检查前缀的最后一位，并且基于该结果，该条目被放置在两个关联存储器条目类之一中。然后，如果前缀的最后一位为零，则通过执行使用前缀填充的前缀的查找操作来识别该条目，或者如果最后一位是填满查找字的剩余位的则为零。不同类别条目的条目通常存储在不同的关联存储器中，或者存储在与每个这些条目相同的关联存储器中，包括类特定的标识符。在其他应用中，这些条目可以用于定位前缀并将树数据结构存储在一个或多个关联存储器中。

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