专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

91. 发明授权

US09098812B2 Faster minimum error rate training for weighted linear models 有权
标题翻译：加权线性模型更快的最小误差率训练
公开(公告)号：US09098812B2
公开(公告)日：2015-08-04
申请号：US12423187
申请日：2009-04-14
申请人： Robert Carter Moore , Christopher Brian Quirk
发明人： Robert Carter Moore , Christopher Brian Quirk
IPC分类号： G06F9/44 , G06N7/02 , G06N7/06 , G06N99/00 , G06N5/04 , G06N5/00
CPC分类号： G06N99/005 , G06N5/003 , G06N5/04
摘要： The claimed subject matter provides systems and/or methods for training feature weights in a statistical machine translation model. The system can include components that obtain lists of translation hypotheses and associated feature values, set a current point in the multidimensional feature weight space to an initial value, chooses a line in the feature weight space that passes through the current point, and resets the current point to optimize the feature weights with respect to the line. The system can further include components that set the current point to be a best point attained, reduce the list of translation hypotheses based on a determination that a particular hypothesis has never been touched in optimizing the feature weights from at least one of an initial staring point or a randomly selected restarting point, and output the point ascertained to be the best point in the feature weight space.
摘要翻译：所要求保护的主题提供用于在统计机器翻译模型中训练特征权重的系统和/或方法。该系统可以包括获得翻译假设和相关特征值的列表的组件，将多维特征权重空间中的当前点设置为初始值，在通过当前点的特征权重空间中选择一行，并重置当前指向相对于线路优化特征权重。该系统可以进一步包括将当前点设定为获得的最佳点的组件，基于在从初始凝视点中的至少一个优化特征权重时从未触及特定假设的确定来减少翻译假设列表或随机选择的重新启动点，并且将确定的点输出为特征权重空间中的最佳点。

92. 发明授权

US09098798B2 Methods and apparatus for prediction and modification of behavior in networks 有权
标题翻译：网络行为预测与修改的方法与装置
公开(公告)号：US09098798B2
公开(公告)日：2015-08-04
申请号：US13482925
申请日：2012-05-29
申请人： Wei Pan , Yaniv Altshuler , Alex Paul Pentland , Nadav Aharony
发明人： Wei Pan , Yaniv Altshuler , Alex Paul Pentland , Nadav Aharony
IPC分类号： G06N7/02 , G06N7/00 , G06Q50/00
CPC分类号： G06N7/005 , G06Q50/01
摘要： In exemplary implementations of this invention, mobile application (app) installations by users of one or more networks are predicted. Using network data gathered by smartphones, multiple “candidate” graphs (including a call log graph) are calculated. The “candidate” graphs are weighted by an optimization vector and then summed to calculate a composite graph. The composite graph is used to predict the conditional probabilities that the respective users will install an app, depending in part on whether the user's neighbors have previously installed the app. Exogenous factors, such as the app's quality, may be taken into account by creating a virtual candidate graph. The conditional probabilities may be used to select a subset of the users. Signals may be sent to the subset of users, including to recommend an app. Also, the probability of successful “trend ignition” may be predicted from network data.
摘要翻译：在本发明的示例性实现中，预测了一个或多个网络的用户的移动应用（app）安装。使用智能手机收集的网络数据，计算出多个“候选”图（包括通话记录图）。 “候选”图由优化向量加权，然后相加以计算合成图。复合图用于预测相应用户将安装应用程序的条件概率，部分取决于用户的邻居是否以前已安装该应用程序。可以通过创建虚拟候选图来考虑外部因素，例如应用程序的质量。条件概率可以用于选择用户的子集。信号可能被发送到用户的子集，包括推荐一个应用程序。此外，可以从网络数据预测成功的“趋势点火”的概率。

93. 发明申请

US20150206059A1 FUZZY INFERENCE DEDUCTION USING RULES AND HIERARCHY-BASED ITEM ASSIGNMENTS 有权
公开(公告)号：US20150206059A1
公开(公告)日：2015-07-23
申请号：US14603152
申请日：2015-01-22
申请人： Melanie Anne McMeekan
发明人： Melanie Anne McMeekan
IPC分类号： G06N5/04 , G06F17/30 , G06N7/02
CPC分类号： G06N5/048 , G06F17/30424 , G06F17/30705 , G06N5/022 , G06N7/02
摘要： Some embodiments disclosed herein relate to generating fuzzy inferences of procedure types based on fuzzy logic. Membership functions can be used to relate item variables to a degree of correspondence to various item types. Fuzzy rules can specify processing to be conducted using membership values produced by evaluations of membership functions. An output of the processing can include an inference that a content object corresponding to the item variables relates to one or more procedure types. Further, some embodiments disclosed herein relate to querying hierarchical data structures to identify related items. A hierarchical data structure can associate each of a set of procedure types with one or more item types and/or item identifiers or characteristics.

94. 发明授权

US09075909B2 System and method to enable detection of viral infection by users of electronic communication devices 有权
标题翻译：用于电子通信设备的用户检测病毒感染的系统和方法
公开(公告)号：US09075909B2
公开(公告)日：2015-07-07
申请号：US13680289
申请日：2012-11-19
申请人： Gal Almogy , Gilad Almogy
发明人： Gal Almogy , Gilad Almogy
IPC分类号： G06F9/44 , G06N7/02 , G06N7/06 , G06F19/00
CPC分类号： G16H50/20 , G06F19/00 , G16H50/80 , Y02A90/24
摘要： A non-transitory computer readable medium that stores instructions for causing a computerized system to perform the following operations: determining, by the computerized system, that a first person is infected by a first infectious disease; wherein the determination is associated with a first person infection probability attribute; detecting, by the computerized system, based upon location information collected during at least a portion of a first infectious disease manifestation period, the location information being indicative of locations of the first person and other persons, a second person that was within an infection distance from the first person and is potentially infected by the first infectious disease; calculating, by the computerized system, a second person infection probability attribute; and updating, by the computerized system, the first person infection probability attribute in response to the second person infection probability attribute.
摘要翻译：存储用于使计算机化系统执行以下操作的指令的非暂时计算机可读介质：由计算机化系统确定第一人被第一感染性疾病感染; 其中所述确定与第一人感染概率属性相关联; 由所述计算机化系统基于在第一感染性疾病表现期间的至少一部分期间收集的位置信息来检测所述位置信息，所述位置信息指示所述第一人和其他人的位置，所述位置信息在距离第一个人并且可能被第一感染性疾病感染; 通过计算机化系统计算第二人感染概率属性; 并且由所述计算机化系统更新响应于所述第二人感染概率属性的所述第一人感染概率属性。

95. 发明授权

US09058811B2 Speech synthesis with fuzzy heteronym prediction using decision trees 有权
标题翻译：使用决策树进行模糊异词预测的语音合成
公开(公告)号：US09058811B2
公开(公告)日：2015-06-16
申请号：US13402602
申请日：2012-02-22
申请人： Xi Wang , Xiaoyan Lou , Jian Li
发明人： Xi Wang , Xiaoyan Lou , Jian Li
IPC分类号： G10L13/02 , G10L13/08 , G06N7/02
CPC分类号： G10L13/08
摘要： According to one embodiment, a method, apparatus for synthesizing speech, and a method for training acoustic model used in speech synthesis is provided. The method for synthesizing speech may include determining data generated by text analysis as fuzzy heteronym data, performing fuzzy heteronym prediction on the fuzzy heteronym data to output a plurality of candidate pronunciations of the fuzzy heteronym data and probabilities thereof, generating fuzzy context feature labels based on the plurality of candidate pronunciations and probabilities thereof, determining model parameters for the fuzzy context feature labels based on acoustic model with fuzzy decision tree, generating speech parameters from the model parameters, and synthesizing the speech parameters via synthesizer as speech.
摘要翻译：根据一个实施例，提供了一种用于合成语音的方法，装置，以及用于语音合成中使用的用于训练声学模型的方法。用于合成语音的方法可以包括通过文本分析产生的数据作为模糊异词数据，对模糊异词数据执行模糊异词预测以输出模糊异词数据的多个候选发音及其概率，基于多个候选发音和概率，基于具有模糊决策树的声学模型确定模糊上下文特征标签的模型参数，从模型参数生成语音参数，并通过合成器将语音参数合成为语音。

96. 发明申请

US20150149391A1 SYSTEM AND A METHOD FOR PRIVIDING A DIALOG WITH A USER 有权
标题翻译：系统和用户隐私对话方法
公开(公告)号：US20150149391A1
公开(公告)日：2015-05-28
申请号：US14092972
申请日：2013-11-28
申请人： Akademia Gorniczo-Hutnicza Im. Stanislawa Staszica W Krakowie
发明人： Bartosz Ziolko , Tomasz Pedzimaz
IPC分类号： G06N7/02 , G06N5/02 , G06N3/00
CPC分类号： G06N7/02 , G06N3/006 , G06N5/02 , G10L13/00 , G10L15/1822 , G10L15/22 , G10L25/63 , G10L2015/227
摘要： A computer-implemented method for providing a dialog with a user, the method comprising the steps of: (a) creating (501) dialog's plot scenario as a narrative graph structure containing the dialogs; (b) assigning (502) coefficients of dialog transitions maps to narrative graph nodes; (c) providing (503) speech input for recognition in a given context and dialog phase represented by current narrative graph node; (d) applying (504) algorithm for updating dialog coefficients based on user's speech or user's other behavior; (e) applying (505) at least one fuzzy logic algorithm which using user speech, and other coefficients on the transition map, determines transition to another narrative graph node (phase of a dialog or a plot), or updates a position on the transition map continuing the dialog in the same narrative graph node (f) determining (507) a response based on coefficients of the narrative graph; (g) repeating steps (c) to (f) for a particular narrative graph node until a transition is decided in step (e); (h) after a transition in step (e) repeating steps (c) to (f) with new coefficients in a new narrative graph node starting with a new position on a dialog graph structure and with new values of coefficients.
摘要翻译：一种用于向用户提供对话的计算机实现的方法，所述方法包括以下步骤：（a）创建（501）对话框的剧情场景作为包含对话的叙述图结构; （b）将（502）对话转换系数映射到叙述图节点; （c）在由当前叙述图节点表示的给定上下文和对话阶段中提供（503）用于识别的语音输入; （d）基于用户的语音或用户的其他行为应用（504）算法来更新对话系数; （e）应用（505）使用用户语音的至少一个模糊逻辑算法和转换图上的其他系数确定向另一个叙述图节点（对话或绘图的阶段）的转变，或者更新转换位置映射在相同叙述图节点中继续对话;（f）基于叙述图的系数确定（507）响应; （g）对于特定叙述图节点重复步骤（c）至（f），直到在步骤（e）中确定转换; （h）在步骤（e）中的转变之后，在新的叙述图节点中以对话框图结构上的新位置和系数的新值开头的新系数重复步骤（c）至（f）。

97. 发明授权

US09037527B2 Determining a dynamic user profile indicative of a user behavior context with a mobile device 有权
标题翻译：确定指示与移动设备的用户行为上下文的动态用户简档
公开(公告)号：US09037527B2
公开(公告)日：2015-05-19
申请号：US14221098
申请日：2014-03-20
申请人： QUALCOMM Incorporated
发明人： Lukas Daniel Kuhn , Sanjiv Nanda , Vidya Narayanan
IPC分类号： G06F9/44 , G06N7/02 , G06N5/04 , H04M1/725 , H04M3/42
CPC分类号： G06N5/048 , H04M1/72569 , H04M3/42136 , H04M2242/30
摘要： Methods, apparatuses and articles of manufacture for use in a mobile device to determine whether a dynamic user profile is to transition from a first state to a second state based, at least in part, on one or more sensed indicators. The dynamic user profile may be indicative of one or more current inferable user behavior contexts for a user co-located with the mobile device. The mobile device may transition a dynamic user profile from a first state to a second state, in response to a determination that the dynamic user profile is to transition from the first state to the second state, and operatively affect one or more functions performed, at least in part, by the mobile device based, at least in part, on the transition of the dynamic user profile to the second state.
摘要翻译：至少部分地基于一个或多个感测到的指示符，在移动设备中使用的方法，装置和制品，以确定动态用户简档是否从第一状态转换到第二状态。动态用户简档可以指示用于与移动设备共处的用户的一个或多个当前可推断的用户行为上下文。响应于确定动态用户简档将从第一状态转换到第二状态，移动设备可以将动态用户简档从第一状态转换到第二状态，并且可操作地影响所执行的一个或多个功能至少部分地由移动设备至少部分地基于动态用户简档向第二状态的转换。

98. 发明授权

US08972330B2 Identification of significant sequences of fault codes by statistical hypothesis testing 有权
标题翻译：通过统计假设检验识别故障代码的重要序列
公开(公告)号：US08972330B2
公开(公告)日：2015-03-03
申请号：US13209583
申请日：2011-08-15
申请人： Shi Zhao
发明人： Shi Zhao
IPC分类号： G06F9/44 , G06N7/02 , G06N7/06 , G06N5/02 , G06Q10/00 , G06F11/07
CPC分类号： G06N5/02 , G06F11/079 , G06Q10/20
摘要： In some aspects of the present application, a method for identifying significant events related to machine problems. The method includes receiving one or more machine problems; determining one or more machine problem classifications based on the one or more machine problems; generating a probability distribution that the one or more machines problems are related to the one more machine problem classifications; determining if one or more events are associated with the one or more machine problems during a predetermined time interval; and determining if the one or more events are significantly related to the one or more machine problem classifications using a statistical algorithm.
摘要翻译：在本申请的一些方面，一种用于识别与机器问题有关的重要事件的方法。该方法包括接收一个或多个机器问题; 基于一个或多个机器问题确定一个或多个机器问题分类; 产生一个或多个机器问题与一个以上机器问题分类相关的概率分布; 在预定时间间隔期间确定一个或多个事件是否与一个或多个机器问题相关联; 以及使用统计算法确定所述一个或多个事件是否与所述一个或多个机器问题分类显着相关。

99. 发明授权

US08972329B2 Systems and methods for ranking nodes of a graph using random parameters 有权
标题翻译：使用随机参数对图的节点进行排序的系统和方法
公开(公告)号：US08972329B2
公开(公告)日：2015-03-03
申请号：US12435244
申请日：2009-05-04
申请人： Paul Constantine , David Gleich
发明人： Paul Constantine , David Gleich
IPC分类号： G06F9/44 , G06N7/02 , G06N7/06 , G06N7/00
CPC分类号： G06F17/3053 , G06F17/30958 , G06N7/005 , G06N99/005
摘要： A ranking approach is used to determine rank-based relationships. In connection with various embodiments, the present invention is directed to a method for ranking nodes of a graph. A vector is provided as a function of a set of random parameters, and a probability matrix function is used, relative to nodes of the graph, to assess the statistics of the vector that solves a probability-based system. Certain embodiments are directed to determining a page rank for a web-based search.
摘要翻译：使用排名方法来确定基于排名的关系。结合各种实施例，本发明涉及一种用于对图表的节点进行排序的方法。根据一组随机参数提供向量，相对于图的节点使用概率矩阵函数来评估解决基于概率的系统的向量的统计。某些实施例涉及确定基于web的搜索的页面排名。

100. 发明授权

US08924337B2 Recursive Bayesian controllers for non-linear acoustic echo cancellation and suppression systems 有权
标题翻译：递归贝叶斯控制器用于非线性声学回声消除和抑制系统
公开(公告)号：US08924337B2
公开(公告)日：2014-12-30
申请号：US13068341
申请日：2011-05-09
申请人： Sarmad Malik , Gerald Enzner , Jukka Petteri Vartiainen , Jari Petteri Sjoberg , Ville Mikael Myllyla
发明人： Sarmad Malik , Gerald Enzner , Jukka Petteri Vartiainen , Jari Petteri Sjoberg , Ville Mikael Myllyla
IPC分类号： G06N7/02
CPC分类号： G06N7/005
摘要： Both a cascade and a multichannel joint Bayesian estimator are provided for suppressing acoustic echo. An expansion basis (Power/Fourier series) is selected to convert a sample-based input signal xt into a DFT-domain multichannel signal [Xτ,1, . . . Xτ,p]. The posterior of unknown states (e.g., mean Ŵτ and covariance Pτ of the echo path Wτ and the mean âτ and covariance Qτ of the nonlinear coefficients aτ; or channel-wise mean Ŵτ,i and multichannel covariance Pτ of a compound quantity formed by merging together the echo path Wτ and the ith nonlinear coefficient aτ,i) and model parameters θτ are estimated; and Kalman gain factor(s) Kτ are computed for optimal adaptation of the posterior of unknown states. An echo signal Ŷτ is estimated using the multichannel input signal [Xτ,1, . . . Xτ,p] and the adapted posterior; and an error signal Eτ is generated. Residual echo is suppressed by post-filtering the error signal Eτ with a weighting function ψτ which depends on the adapted posterior, and the filtered error signal ŝ′t is then transmitted to a far-end.
摘要翻译：提供级联和多通道联合贝叶斯估计器来抑制声学回声。选择扩展基础（功率/傅立叶级数）以将基于采样的输入信号xt转换为DFT域多通道信号[Xτ，1，...，。。 Xτ，p]。未知状态的后验（例如，回波路径Wτ的平均值和协方差Pτ以及非线性系数aτ的平均值和协方差Qτ或通过合并形成的复合数量的通道均值Ŵτ，i和多通道协方差Pτ 一起估计回波路径Wτ和第i个非线性系数aτ，i）和模型参数τ; 和卡尔曼增益因子（K）被计算用于未知状态的后验的最佳适应。使用多通道输入信号[Xτ，1，...]来估计回波信号Ŷτ。。。 Xτ，p]和适应的后部; 并产生误差信号Eτ。通过对取决于适合的后验的加权函数ψτ对误差信号Eτ进行后滤波来抑制残留回波，然后将滤波后的误差信号发送到远端。

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