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

US20050273317A1 Method and apparatus for unsupervised training of natural language processing units 有权
标题翻译：自然语言处理单元无人训练的方法和装置
公开(公告)号：US20050273317A1
公开(公告)日：2005-12-08
申请号：US11204213
申请日：2005-08-15
申请人： Eric Brill , Arul Menezes
发明人： Eric Brill , Arul Menezes
IPC分类号： G06F17/27
CPC分类号： G06F17/274
摘要： A method of training a natural language processing unit applies a candidate learning set to at least one component of the natural language unit. The natural language unit is then used to generate a meaning set from a first corpus. A second meaning set is generated from a second corpus using a second natural language unit and the two meaning sets are compared to each other to form a score for the candidate learning set. This score is used to determine whether to modify the natural language unit based on the candidate learning set.
摘要翻译：训练自然语言处理单元的方法将候选学习集合应用于自然语言单元的至少一个分量。然后，自然语言单元用于从第一语料库生成意义集。使用第二自然语言单元从第二语料库生成第二含义集合，并且将两个含义集合彼此进行比较以形成候选学习集合的分数。该分数用于确定是否基于候选学习集修改自然语言单元。

2. 发明申请

US20050228640A1 Statistical language model for logical forms 有权
标题翻译：逻辑形式的统计语言模型
公开(公告)号：US20050228640A1
公开(公告)日：2005-10-13
申请号：US10813208
申请日：2004-03-30
申请人： Anthony Aue , Eric Ringger , Christopher Quirk , Arul Menezes , Robert Moore
发明人： Anthony Aue , Eric Ringger , Christopher Quirk , Arul Menezes , Robert Moore
IPC分类号： G06F17/27 , G06F17/28
CPC分类号： G06F17/2818 , G06F17/2881
摘要： A method of decoding an input semantic structure to generate an output semantic structure. A set of transfer mappings are provided. A score is calculated for at least one transfer mapping in the set of transfer mappings using a statistical model. At least one transfer mapping is selected based on the score and used to construct the output semantic structure.
摘要翻译：一种解码输入语义结构以生成输出语义结构的方法。提供一组传输映射。使用统计模型计算一组传输映射中的至少一个传输映射的分数。基于分数选择至少一个传输映射，并用于构建输出语义结构。

3. 发明授权

US5742825A Operating system for office machines 失效
标题翻译：办公机器操作系统
公开(公告)号：US5742825A
公开(公告)日：1998-04-21
申请号：US803845
申请日：1997-02-28
申请人： Sharad Mathur , Franklin D. Fite, Jr. , Arul Menezes , Kim Stebbens
发明人： Sharad Mathur , Franklin D. Fite, Jr. , Arul Menezes , Kim Stebbens
IPC分类号： G06F15/00 , G06F9/46 , G06F9/48 , G06F9/40
CPC分类号： G06F9/4887 , G06F9/546
摘要： A multitasking, graphical windowing operating system for an office machine supporting real time processing, including a method for supporting real time processing in the operating system. The operating system includes a kernel having a non-preemptive scheduler for scheduling windowing applications and a preemptive scheduler for scheduling real time applications. The kernel includes a messaging subsystem for communication among process in the system including both the real time and windowing applications, and further includes an event semaphore maintained by the kernel to manage scheduling. The method includes grouping windowing applications together for scheduling purposes and scheduling windowing applications non-preemptively. The method further includes determining priority of real time applications and scheduling real time applications preemptively with the windowing applications based on priority.
摘要翻译：一种用于支持实时处理的办公机器的多任务图形窗口操作系统，包括用于在操作系统中支持实时处理的方法。操作系统包括具有用于调度加窗应用的非抢占式调度器的内核和用于调度实时应用的抢占式调度器。内核包括用于在系统中的进程之间进行通信的消息传递子系统，包括实时应用和开窗应用，并且还包括由内核维护以管理调度的事件信号量。该方法包括将加窗应用程序组合在一起用于调度目的，并且非抢先地调度开窗应用。该方法进一步包括确定实时应用的优先级，并基于优先级预先与开窗应用程序调度实时应用。

4. 发明申请

US20080154848A1 Search, Analysis and Comparison of Content 有权
标题翻译：内容搜索，分析与比较
公开(公告)号：US20080154848A1
公开(公告)日：2008-06-26
申请号：US11613958
申请日：2006-12-20
申请人： Brian Dean Haslam , Patrick Wayne John Evans , Arul Menezes , Patrick Joseph Dinio Santos
发明人： Brian Dean Haslam , Patrick Wayne John Evans , Arul Menezes , Patrick Joseph Dinio Santos
IPC分类号： G06F17/30
CPC分类号： G06F17/30011 , G06F17/30728 , G06F2216/11
摘要： The present invention may be used to analyze subject content, search and analyze reference content, compare the subject and reference content for similarity, and output comparison reports between the subject and reference content. The present invention may incorporate and utilize text from intrinsic and/or extrinsic subject documents. The analysis may employ a variety of metrics, including scores generated from a natural language processing system, scores based on classification similarity, scores based on proximity similarity, and in the case of analysis of patent documents, scores based on measurement of claims.
摘要翻译：本发明可以用于分析主题内容，搜索和分析参考内容，比较主题和参考内容的相似性，并且输出主题和参考内容之间的比较报告。本发明可以包含和利用来自内在和/或外在主题文件的文本。分析可以采用各种度量，包括从自然语言处理系统产生的分数，基于分类相似度的分数，基于邻近度相似度的分数，以及在专利文献分析的情况下，基于权利要求的测量的分数。

5. 发明申请

US20060111896A1 Projecting dependencies to generate target language dependency structure 有权
标题翻译：投影依赖关系来生成目标语言依赖结构
公开(公告)号：US20060111896A1
公开(公告)日：2006-05-25
申请号：US11014108
申请日：2004-12-16
申请人： Arul Menezes , Christopher Quirk , Colin Cherry
发明人： Arul Menezes , Christopher Quirk , Colin Cherry
IPC分类号： G06F17/21
CPC分类号： G06F17/2818 , G06F17/28
摘要： In one embodiment of the present invention, a decoder receives a dependency tree as a source language input and accesses a set of statistical models that produce outputs combined in a log linear framework. The decoder also accesses a table of treelet translation pairs and returns a target dependency tree based on the source dependency tree, based on access to the table of treelet translation pairs, and based on the application of the statistical models.
摘要翻译：在本发明的一个实施例中，解码器接收依赖树作为源语言输入，并访问产生在对数线性框架中组合的输出的一组统计模型。解码器还访问树形图转换对的表，并基于对依赖树的转换对的访问，并且基于统计模型的应用，基于源依赖关系树返回目标依赖关系树。

6. 发明授权

US5499109A System for transferring messages between input and output devices in a communication device 失效
公开(公告)号：US5499109A
公开(公告)日：1996-03-12
申请号：US198364
申请日：1994-02-18
申请人： Sharad Mathur , Arul Menezes , Michael R. VanBuskirk , Kevin T. Gallo
发明人： Sharad Mathur , Arul Menezes , Michael R. VanBuskirk , Kevin T. Gallo
IPC分类号： H04L12/58 , H04L29/06 , H04N1/32 , H04N1/00
CPC分类号： H04L12/5835 , H04L51/066 , H04L69/08 , H04N1/32566 , H04N2201/0065
摘要： A method and system for processing messages is provided. In a preferred embodiment of the present invention, a message scheduler, a job process, and a format resolution module are used. The message scheduler identifies when a new message is to be processed, schedules the new message for processing, invokes the format resolution module to determine a method of converting data in the message to an appropriate format for a destination device or file, and invokes a job process. The format resolution module determines an appropriate method for converting data in the message from a format acceptable to the source devices or files to a format acceptable to the destination devices or files and creates a target message containing a reference to a destination device or file. The job process sends the data between the source devices or files to a destination device or file and performs the appropriate conversion. The preferred embodiment of the present invention is designed to allow for the support of new devices and new data formats with little change to the existing system through the use of a building block methodology. Therefore, new devices and data formats can be supported in an easy and efficient manner.

7. 发明授权

US08065307B2 Parsing, analysis and scoring of document content 有权
标题翻译：解析，分析和评分文件内容
公开(公告)号：US08065307B2
公开(公告)日：2011-11-22
申请号：US11613958
申请日：2006-12-20
申请人： Brian Dean Haslam , Patrick Wayne John Evans , Arul Menezes , Patrick Santos
发明人： Brian Dean Haslam , Patrick Wayne John Evans , Arul Menezes , Patrick Santos
IPC分类号： G06F17/30
CPC分类号： G06F17/30011 , G06F17/30728 , G06F2216/11
摘要： The present invention may be used to analyze subject content, search and analyze reference content, compare the subject and reference content for similarity, and output comparison reports between the subject and reference content. The present invention may incorporate and utilize text from intrinsic and/or extrinsic subject documents. The analysis may employ a variety of metrics, including scores generated from a natural language processing system, scores based on classification similarity, scores based on proximity similarity, and in the case of analysis of patent documents, scores based on measurement of claims.
摘要翻译：本发明可以用于分析主题内容，搜索和分析参考内容，比较主题和参考内容的相似性，并且输出主题和参考内容之间的比较报告。本发明可以包含和利用来自内在和/或外在主题文件的文本。分析可以采用各种度量，包括从自然语言处理系统产生的分数，基于分类相似度的分数，基于邻近度相似度的分数，以及在专利文献分析的情况下，基于权利要求的测量的分数。

8. 发明申请

US20070255550A1 Machine translation split between front end and back end processors 有权
标题翻译：机器翻译分为前端和后端处理器
公开(公告)号：US20070255550A1
公开(公告)日：2007-11-01
申请号：US11414844
申请日：2006-05-01
申请人： Arul Menezes , Christopher Quirk , Richard Rashid , Robert Moore
发明人： Arul Menezes , Christopher Quirk , Richard Rashid , Robert Moore
IPC分类号： G06F17/28
CPC分类号： G06F17/289
摘要： A method of translation includes uploading a source text portion to a back end processor. The back end processor identifies a subset of translation knowledge associated with the source text portion. The back end processor downloads the subset to a front end processor. A translation engine runs on the front end processor. The translation engine generates a translation of the source text portion as a function of the subset.
摘要翻译：一种翻译方法包括将源文本部分上传到后端处理器。后端处理器识别与源文本部分相关联的翻译知识的子集。后端处理器将子集下载到前端处理器。翻译引擎在前端处理器上运行。翻译引擎生成作为子集的函数的源文本部分的翻译。

9. 发明申请

US20050065777A1 System and method for matching a textual input to a lexical knowledge based and for utilizing results of that match 失效
标题翻译：用于将文本输入与基于词汇知识相匹配并用于利用该匹配的结果的系统和方法
公开(公告)号：US20050065777A1
公开(公告)日：2005-03-24
申请号：US10977910
申请日：2004-10-29
申请人： William Dolan , Michael Barnett , Stephen Richardson , Arul Menezes , Lucretia Vanderwende
发明人： William Dolan , Michael Barnett , Stephen Richardson , Arul Menezes , Lucretia Vanderwende
IPC分类号： G06F17/27 , G06F17/30 , G06F17/21
CPC分类号： G06F17/30684 , G06F17/271 , G06F17/277 , G06F17/2785 , Y10S707/99932 , Y10S707/99935
摘要： The present invention can be used in a natural language processing system to determine a relationship (such as similarity in meaning) between two textual segments. The relationship can be identified or determined based on logical graphs generated from the textual segments. A relationship between first and second logical graphs is determined. This is accomplished regardless of whether there is an exact match between the first and second logical graphs. In one embodiment, the first graph represents an input textual discourse unit. The second graph, in one embodiment, represents information in a lexical knowledge base (LKB). The input graph can be matched against the second graph, if they have similar meaning, even if the two differ lexically or structurally.
摘要翻译：本发明可以用于自然语言处理系统中以确定两个文本段之间的关系（诸如意义上的相似性）。可以基于从文本段生成的逻辑图来识别或确定关系。确定第一和第二逻辑图之间的关系。无论第一个和第二个逻辑图之间是否存在精确的匹配，这是完成的。在一个实施例中，第一图表示输入文本话语单元。在一个实施例中，第二个图表示词汇知识库（LKB）中的信息。输入图可以与第二个图匹配，如果它们具有相似的含义，即使两者在词汇或结构上不同。

10. 发明授权

US5473691A System and method for computer data transmission 失效
标题翻译：计算机数据传输的系统和方法
公开(公告)号：US5473691A
公开(公告)日：1995-12-05
申请号：US149265
申请日：1993-11-05
申请人： Arul Menezes , Sharad Mathur , Michael Ginsberg
发明人： Arul Menezes , Sharad Mathur , Michael Ginsberg
IPC分类号： G06F13/00 , H04L29/06 , H04L29/10 , H04N1/00 , H04N1/32 , H04K1/00
CPC分类号： H04N1/32128 , H04L29/06 , H04N1/32112 , H04N2201/3204 , H04N2201/3205 , H04N2201/3207 , H04N2201/3233 , H04N2201/3243 , H04N2201/3278
摘要： The system of the present invention processes data for communication between first and second computers by linearizing the communications message. The message comprises a linear header portion, an extended header portion, and a message body. The linear header portion identifies the number of message recipients and message types. The extended header contains detailed information about the message recipients, such as recipient name and address. The extended header may also contain message subject information, polling information, and password data. The header information is used by the receiving computer to prepare to process the expected data type. The message body may be transferred in a form that takes advantage of the data processing capabilities of the first and second computers. The computers may exchange data processing capabilities so that the most efficient transfer form may be selected. The extended header and message body are encoded using a well-known ASN-1 data encoding process. In addition, the message body may be compressed and encrypted. The system may be readily used in facsimile communication where the first and second computers are facsimile machines.
摘要翻译：本发明的系统通过线性化处理通信消息来处理第一和第二计算机之间的通信的数据。消息包括线性报头部分，扩展报头部分和消息体。线性标题部分标识消息收件人和消息类型的数量。扩展头包含有关邮件收件人的详细信息，如收件人姓名和地址。扩展报头还可以包含消息对象信息，轮询信息和密码数据。接收计算机使用头信息来准备处理期望的数据类型。消息体可以以利用第一和第二计算机的数据处理能力的形式传送。计算机可以交换数据处理能力，以便可以选择最有效的传送形式。使用公知的ASN-1数据编码处理来对扩展的报头和消息体进行编码。另外，消息体可以被压缩和加密。该系统可以容易地用于传真通信，其中第一和第二计算机是传真机。

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