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. 发明专利

JP2008077163A Search system, search method and search program 有权
标题翻译：搜索系统，搜索方法和搜索程序
公开(公告)号：JP2008077163A
公开(公告)日：2008-04-03
申请号：JP2006252592
申请日：2006-09-19
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： YOSHIMURA HIROKI , UCHIHASHI SHINGO
IPC分类号： G06F17/30 , G06Q50/22 , G06Q50/24 , G06T1/00
摘要： PROBLEM TO BE SOLVED: To provide a search system, a search method and a search program that can produce search results simply without requiring operations such as keyword input. SOLUTION: A PC comprising the search system has a search part 12 for inputting new diagnostic image information and calculating a similarity between the new diagnostic image and past diagnostic image information in each interpretation report data structure, an analysis part 14 for calculating the occurrence rate of a diagnosis name included in past interpretation report information in each interpretation report data structure, and an examination part 16 for calculating a comprehensive decision value representing fitness for a search result from the similarity and occurrence rate and acquiring a predetermined number of pieces of past diagnostic image information and pieces of past interpretation report information corresponding to the pieces of past diagnostic image information in descending order of the comprehensive decision value as search results of past diagnostic image information similar to the new diagnostic image information and past interpretation report information corresponding to the past diagnostic image information. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：提供可以简单地不需要诸如关键字输入的操作来产生搜索结果的搜索系统，搜索方法和搜索程序。解决方案：包括搜索系统的PC具有搜索部分12，用于输入新的诊断图像信息，并且在每个解释报告数据结构中计算新诊断图像和过去的诊断图像信息之间的相似度，分析部分14用于计算每个解释报告数据结构中的过去的解释报告信息中包括的诊断名称的发生率，以及用于从相似度和发生率计算表示搜索结果的适合度的综合判定值的检查部分16，并且获取预定数量的过去的诊断图像信息和与过去诊断图像信息相对应的过去诊断图像信息的过去的诊断图像信息和综合判定值的降序作为与新诊断图像信息和过去解释报告信息对应的过去诊断图像信息的搜索结果对过去的诊断图像信息进行处理。版权所有（C）2008，JPO＆INPIT

2. 发明专利

JP2007207127A Question answering system, question answering processing method and question answering program 审中-公开
标题翻译：问题回答系统，问题解答处理方法和问题解答程序
公开(公告)号：JP2007207127A
公开(公告)日：2007-08-16
申请号：JP2006027804
申请日：2006-02-04
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： MASUICHI HIROSHI , YOSHIOKA TAKESHI , YOSHIMURA HIROKI , SUGIHARA DAIGO , OKUMA TOMOKO
IPC分类号： G06F17/30
摘要： PROBLEM TO BE SOLVED: To provide a question answering system, a question answering processing method and a question answering program, reducing man-hours of description of a conversion rule by generating the proper conversion rule according to a retrieval question.
SOLUTION: This question answering system 100 has: a similar question generation part 40 generating a similar retrieval question sentence wherein a characteristic term included in a retrieval question sentence is replaced with a similar term having the same attribute as the characteristic term; an expansion term retrieval part 50 acquiring a retrieval result sentence including an answer to the similar retrieval question sentence, the similar term, and things inquired by the similar retrieval question sentence; a character string pattern generation part 60 acquiring arrangement of the answer, the similar term, and the things, in the sentence; and a retrieval part 70 extracting a word in the retrieval result sentence corresponding to the answer to the similar retrieval question sentence as an answer on the basis of the arrangement.
COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：提供问答系统，问答处理方法和问答程序，通过根据检索问题生成适当的转换规则来减少对转换规则的描述的工时。该问题回答系统100具有：类似的问题生成部分40，生成类似的检索问题，其中包括在检索问题句中的特征项被替换为与特征项具有相同属性的类似项; 扩展项检索部分50获取包括类似检索问题的答案，类似术语和由类似的检索问题询问的事物的检索结果句子; 字符串图案生成部60，取得句子中的答案，类似词语和事物的排列; 以及检索部70，基于该配置，将与相似的检索问题的答案相对应的检索结果句中的单词提取为答案。版权所有（C）2007，JPO＆INPIT

3. 发明专利

JP2006279098A Translation apparatus, translation method and program 有权
标题翻译：翻译设备，翻译方法和程序
公开(公告)号：JP2006279098A
公开(公告)日：2006-10-12
申请号：JP2005090198
申请日：2005-03-25
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： KATO MASAHIRO , IKEGAMI HIROAKI , ITONORI KATSUHIKO , ONDA MASANORI , ASHIKAGA HIDEAKI , KIMURA SHUNICHI , SATAKE MASAKI , YOSHIMURA HIROKI
IPC分类号： H04N1/387 , G06F17/28
CPC分类号： G06F17/289 , G06F21/608 , G06F2221/0737 , H04N1/00204 , H04N1/32203 , H04N2201/0091 , H04N2201/3222 , H04N2201/3271
摘要： PROBLEM TO BE SOLVED: To provide a technology capable of informing a user about the necessity of encryption when a document is to be encrypted. SOLUTION: A translation apparatus receives data for representing a document and extracts security information being an image representing that the document is to be encrypted from the data. When extracting the security information, the translation apparatus translates the security information into a target language and outputs the translated security information. COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：提供能够在要加密文档时通知用户加密必要性的技术。解决方案：翻译装置接收用于表示文档的数据，并从数据中提取表示要加密文档的图像的安全信息。当提取安全信息时，翻译装置将安全信息转换为目标语言，并输出已翻译的安全信息。版权所有（C）2007，JPO＆INPIT

4. 发明专利

JP2006252047A Language processor, and language processing program 审中-公开
标题翻译：语言处理程序和语言处理程序
公开(公告)号：JP2006252047A
公开(公告)日：2006-09-21
申请号：JP2005065959
申请日：2005-03-09
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： SUGIHARA DAIGO , MASUICHI HIROSHI , KIMURA SHUNICHI , ITONORI KATSUHIKO , ASHIKAGA HIDEAKI , YOSHIMURA HIROKI , ONDA MASANORI , KATO MASAHIRO , SATAKE MASAKI
IPC分类号： G06F17/27
CPC分类号： G06F17/279
摘要： PROBLEM TO BE SOLVED: To prevent erroneous analysis on correspondence analysis in a case where the pause of a correspondence link occurs with an antecedent part concerning the fragment of a text being a processing object. SOLUTION: A text input means 11 acquires text data. A document part determining means 15 determines whether the pause of the correspondence link exists in the text data which is acquired by the text input means 11. A correspondence analysis processing means 17 performs correspondence analysis processing in the text data in response to reference which differs by presence or absence of the pause of the correspondence link determined by the document part determining means 15 as the reference for selecting an antecedent candidate from the text data. Then the processed text data is outputted. COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：为了防止在与作为处理对象的文本的片段的先前部分发生对应关系的暂停的情况下的对应分析的错误分析。文本输入装置11获取文本数据。文档部分确定装置15确定对应链接的暂停是否存在于由文本输入装置11获取的文本数据中。对应分析处理装置17响应于参考而进行文本数据中的对应分析处理，存在或不存在由文档部分确定装置15确定的对应链接的暂停作为用于从文本数据中选择先行候选者的参考。然后输出处理后的文本数据。版权所有（C）2006，JPO＆NCIPI

5. 发明专利

JP2006119697A Question answering system, question answering method, and question answering program 审中-公开
标题翻译：问题回答系统，问题解答方法和问题解答程序
公开(公告)号：JP2006119697A
公开(公告)日：2006-05-11
申请号：JP2004303884
申请日：2004-10-19
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： YOSHIMURA HIROKI , MASUICHI HIROSHI , OKUMA TOMOKO , TAKAAI MOTOYUKI , SUGIHARA DAIGO , YAMAZAWA MIYUKI
IPC分类号： G06F17/30
摘要： PROBLEM TO BE SOLVED: To provide a question answering system capable of improving the precision of an answer by improving the precision of information retrieval by precisely specifying a question type, i.e. using a result of question type detection for the information retrieval and the precision of specification of the question type by using an information retrieval result for the question type detection.
SOLUTION: The question answering system 100 comprises: a question analyzing means 110 of analyzing an inputted question sentence and outputting analytic information; an information retrieving means 120 of retrieving information from an information source 130 based upon the analytic information; a question type detecting means 140 of detecting the type of the question sentence based upon the analytic information; a feedback control means 200 of designating feedback processing of the question type detecting means 140 or information retrieving means 120 based upon the information retrieval result or question type detection result; and an answer extracting means 150 of extracting an answer candidate for the question sentence based upon the information retrieval result or question type detection result having been fed back.
COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：提供一种能够通过精确指定问题类型来提高信息检索的精度来提高答案精度的问答系统，即，使用问题类型检测的结果用于信息检索和通过使用问题类型检测的信息检索结果对问题类型的规范的精度。问题解答系统100包括：问题分析装置110，分析输入的问题并输出分析信息; 信息检索装置120，用于根据分析信息从信息源130检索信息; 一种基于分析信息检测问题类型的问题类型检测装置140; 基于信息检索结果或问题类型检测结果指定问题类型检测装置140或信息检索装置120的反馈处理的反馈控制装置200; 以及答辩提取装置150，其基于已经反馈的信息检索结果或问题类型检测结果提取问题的答复候选。版权所有（C）2006，JPO＆NCIPI

6. 发明专利

JP2006065508A Information processing system, information processing method, and computer program 审中-公开
标题翻译：信息处理系统，信息处理方法和计算机程序
公开(公告)号：JP2006065508A
公开(公告)日：2006-03-09
申请号：JP2004245681
申请日：2004-08-25
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： OKUMA TOMOKO , YOSHIOKA TAKESHI , HAYASHI KAZUTAKA , YOSHIMURA HIROKI
IPC分类号： G06F17/30 , G06F17/27 , G06Q10/00 , G06Q10/06 , G06Q50/00
摘要： PROBLEM TO BE SOLVED: To recognize the activities of workers engaged in intellectual labor by classifying the texts described by the workers according to different activities that constitute practices. SOLUTION: A syntactic and semantic analysis process is applied to natural language texts such as texts written by workers engaged in intellectual labor so as to obtain the results of syntactic and semantic analysis such as verbs and case relations, and the aspects and tenses of the verbs. The results of the syntactic and semantic analysis are searched for attribute values that express different classifications of the practices. Based on the attribute values obtained, the natural language texts are classified into different kinds of practice activities, and important words about "What" and "How" are extracted on the basis of the attribute names of the attribute values obtained. COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：通过根据构成实践的不同活动对工人所描述的案例进行分类，来认识从事智力劳动的工作者的活动。解决方案：句法和语义分析过程被应用于自然语言文本，例如从事智力劳动的工作人员撰写的文本，以获得句法和语义分析的结果，如动词和案例关系，以及方面和时态的动词。搜索句法和语义分析的结果，对表示不同分类实践的属性值进行搜索。根据所获得的属性值，将自然语言文本分为不同类型的练习活动，并根据所获得的属性值的属性名称提取“What”和“How”等重要词。版权所有（C）2006，JPO＆NCIPI

7. 发明专利

JP2006059049A Information search system, method, and program 有权
标题翻译：信息搜索系统，方法和程序
公开(公告)号：JP2006059049A
公开(公告)日：2006-03-02
申请号：JP2004239098
申请日：2004-08-19
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： SUGIHARA DAIGO , MASUICHI HIROSHI , YOSHIOKA TAKESHI , HAYASHI KAZUTAKA , OKUMA TOMOKO , YOSHIMURA HIROKI
IPC分类号： G06F17/30
摘要： PROBLEM TO BE SOLVED: To provide an information search system more flexibly comparing a search query sentence with a search objective sentence. SOLUTION: This information search system is provided with a search query sentence input part 206 inputting the search query sentence, a syntax meaning analysis part 202 performing syntax meaning analysis on the search objective sentence depicted by a natural language sentence and the search query sentence, a document structurization part 203 creating a structured sentence from the search objective sentence on which the syntax meaning analysis is carried out, a search formula creation part 207 creating a first search formula matching the search query sentence based on the result of the syntax meaning analysis, a search execution part 209 performing search in the structured document by using the first search formula, and a search result regulation part 211 determining whether there is need to carry out search again or not from the search result. The search formula creation part 207 creates a second search formula having the lower degree of correspondence to the syntax meaning analysis result on the search query sentence in comparison with that by the first search formula, and the search execution part 207 carries out search again on the structured document by using the second search formula. COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：为了使搜索查询句与搜索目标句更灵活地提供信息搜索系统。解决方案：该信息搜索系统具有输入搜索查询语句的搜索查询语句输入部分206，语法意义分析部分202，对由自然语言句子和搜索查询描绘的搜索目标句子进行语法意义分析句子，从执行语法意义分析的搜索目标句子创建结构化句子的文档结构化部分203，基于语法意义的结果创建与搜索查询句子匹配的第一搜索公式的搜索公式创建部分207 分析，搜索执行部分209通过使用第一搜索公式执行结构化文档中的搜索，以及搜索结果规定部分211，确定是否需要从搜索结果再次执行搜索。与第一搜索公式相比，搜索公式创建部207创建与搜索查询语句的语法意义分析结果对应度较小的第二搜索公式，并且搜索执行部207再次对搜索结果进行搜索结构化文档通过使用第二个搜索公式。版权所有（C）2006，JPO＆NCIPI

8. 发明专利

JP2006031528A Device and system for managing document 有权
标题翻译：用于管理文档的设备和系统
公开(公告)号：JP2006031528A
公开(公告)日：2006-02-02
申请号：JP2004211465
申请日：2004-07-20
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： IKEGAMI HIROAKI , ITONORI KATSUHIKO , KATO MASAHIRO , KIMURA SHUNICHI , YOSHIMURA HIROKI , ASHIKAGA HIDEAKI , SATAKE MASAKI
IPC分类号： G06F17/30 , G06Q10/00 , G06Q10/06 , G06Q50/00
CPC分类号： G06F17/30011 , G06Q10/06 , G06Q10/087 , G06Q20/203 , H04N1/00342 , H04N1/2166 , H04N1/32122 , H04N2201/3226 , H04N2201/3254 , H04N2201/327 , H04N2201/3274
摘要： PROBLEM TO BE SOLVED: To provide a device and a system for managing documents which selects documents which are better to be electronized out of documents used in an office, etc. and sets priority for electronization to each document. SOLUTION: A document ID, a use place and a use time about the used documents 10 are stored in a storage device 60. A server 5 analyzes information stored in the storage device 60 and use efficiency of document data to be obtained by digitizing the document 10 by every document 10. The server 50 specifies a document 10 corresponding to document data with high use efficiency by comparing use efficiency of each piece of calculated document data and outputs a specification result to a client 70. In addition, the server 50 sets the priority for digitization to each document 10 on the basis of a comparison result of the use efficiency of each piece of document data and outputs the priority to the client 70. COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：提供一种用于管理文件的设备和系统，其从在办公室等中使用的文档中选择更好地被电子化的文档，并且为每个文档设置电子化的优先级。解决方案：将关于使用的文档10的文档ID，使用地点和使用时间存储在存储装置60中。服务器5分析存储在存储装置60中的信息，并且通过以下方式获得文档数据的效率：通过每个文档10对文档10进行数字化。服务器50通过比较每个计算的文档数据的使用效率来指定对应于具有高效率的文档数据的文档10，并将规范结果输出到客户端70.另外，服务器根据每条文件数据的使用效率的比较结果，将每个文档10的数字化优先级设置为每个文档10的优先级，并将优先级输出给客户端70.版权所有：（C）2006，JPO＆NCIPI

9. 发明专利

JP2005354291A Image reading system, instruction entry apparatus, image reading apparatus, image reading method, and program 审中-公开
标题翻译：图像读取系统，指令录入装置，图像读取装置，图像读取方法和程序
公开(公告)号：JP2005354291A
公开(公告)日：2005-12-22
申请号：JP2004171491
申请日：2004-06-09
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： SATAKE MASAKI , KIMURA SHUNICHI , KATO MASAHIRO , IKEGAMI HIROAKI , YOSHIMURA HIROKI , ITONORI KATSUHIKO , ASHIKAGA HIDEAKI
IPC分类号： G06F3/12 , G06K17/00 , H04N1/00
摘要： PROBLEM TO BE SOLVED: To facilitate settings of an image reading apparatus depending on its model or apparatus body. SOLUTION: An instruction entry apparatus 10 has a conversion table wherein instructions entered to the instruction entry apparatus 10 and codes generated by a code system independently of the image reading apparatus are stored in cross-reference with each other. When an instruction to be instructed to the image reading apparatus 30 is entered to the instruction entry apparatus 10, the instruction entry apparatus 10 extracts the code in cross-reference with the instruction from the conversion table and outputs the extracted code. The image reading apparatus 30 has an inverse conversion table wherein codes outputted from the instruction entry apparatus 10 and instructions from which output results according to instructions entered to the instruction entry apparatus 10 are obtained by the image reading apparatus 30 are stored in cross-reference with each other. The image reading apparatus 30 acquires the code outputted from the instruction entry apparatus 10 from the inverse conversion table, and reads an original and generates original image data denoting the image of the original on the basis of the extracted instruction. COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：为了便于根据其型号或装置主体设置图像读取装置。解决方案：指令输入装置10具有转换表，其中输入到指令输入装置10的指令和与图像读取装置独立的代码系统生成的代码彼此交叉参考。当指示进入图像读取装置30的指令进入指示输入装置10时，指令输入装置10以与转换表的指示相交的方式提取代码，并输出提取的代码。图像读取装置30具有逆转换表，其中从指示输入装置10输出的代码和根据输入到指令输入装置10的指令的输出结果的指令由图像读取装置30获得，与彼此。图像读取装置30从反转换表取得从指示输入装置10输出的代码，并根据提取出的指令，读取原稿并生成表示原稿图像的原始图像数据。版权所有（C）2006，JPO＆NCIPI

10. 发明专利

JP2005092254A Natural language processing system, natural language processing method and computer program 有权
标题翻译：自然语言处理系统，自然语言处理方法和计算机程序
公开(公告)号：JP2005092254A
公开(公告)日：2005-04-07
申请号：JP2003320327
申请日：2003-09-11
申请人： Fuji Xerox Co Ltd , 富士ゼロックス株式会社
发明人： MASUICHI HIROSHI , OKUMA TOMOKO , YOSHIMURA HIROKI , SUGIHARA DAIGO
IPC分类号： G06F17/27
摘要： PROBLEM TO BE SOLVED: To reduce cost necessary for analysis accompanying application of vast rules for imparting semantic information. SOLUTION: A semantic analytical rule is classified into a rule requiring reference to a parsing tree and a rule requiring no reference. Semantic attribute information decidable from a line of morpheme information obtained by morpheme analysis is previously specified, and the obtained semantic attribute information is imparted to a morpheme analysis result in a form not affecting a parsing process and is transferred to parsing. Even when a large number of parsing result candidates are present, it is not necessary to apply the semantic analysis rule to all the candidates, to reduce the calculation cost necessary for the analysis. COPYRIGHT: (C)2005,JPO&NCIPI
摘要翻译：要解决的问题：为了减少用于分配语义信息的广泛规则的分析所需的成本。解决方案：语义分析规则分为需要引用解析树的规则和不需要引用的规则。先前规定了通过语素分析得到的语素信息中可以确定的语义属性信息，并将获得的语义属性信息以不影响解析过程的形式传递给语素分析结果，并转移到语法分析。即使存在大量解析结果候选者，也不需要将语义分析规则应用于所有候选者，以减少分析所需的计算成本。版权所有（C）2005，JPO＆NCIPI

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