专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

11. 发明专利

JP2004326480A Distributed parallel analysis method of mass data 审中-公开
公开(公告)号：JP2004326480A
公开(公告)日：2004-11-18
申请号：JP2003120900
申请日：2003-04-25
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： OKOCHI KAZUYA , MAKI HIDEYUKI , MORITA TOYOHISA
IPC分类号： G06F19/00 , G06N5/04
摘要： PROBLEM TO BE SOLVED: To solve the problem wherein an algorithm does not stop after practical time, when the scale of the problem increases in a method subject to a conventional single processor, and that large a memory area is needed, in a distributed parallel processing method on data mining which finds out knowledge from a large volume of data. SOLUTION: The system conducts analysis, using one analysis management device, one data storing device, and one or more analysis execution devices. The analysis management device manages the analysis execution device and parameters for effective analysis. Also, the analysis management device and the analysis execution device conduct analysis, each maintaining a feature rule whose evaluation value is high, and saves the quantity of the memory required. COPYRIGHT: (C)2005,JPO&NCIPI

12. 发明专利

JP2004139199A Data analysis method 审中-公开
标题翻译：数据分析方法
公开(公告)号：JP2004139199A
公开(公告)日：2004-05-13
申请号：JP2002301225
申请日：2002-10-16
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： MAKI HIDEYUKI , MORITA TOYOHISA , TOMITA HIROYUKI , ITO YUKIYASU
IPC分类号： C12N15/09 , C12Q1/68 , G06F17/15 , G06F17/18 , G06F17/30
摘要： PROBLEM TO BE SOLVED: To solve the problem wherein although the introduction of a DNA microarray enables simultaneous measurement of expression data on many genes, an analysis technique that handles such a large amount of data is ineffective.
SOLUTION: A data analysis method comprises a step of selecting descriptive attributes and a target attribute from among a plurality of attributes in data to be analyzed, a step of analyzing relations between the descriptive attributes and the target attribute and saving the results, a step of presenting the analysis results to a user, a step of selecting one or a plurality of attributes specified by the user from among the descriptive attributes included in the analysis results, and a step of removing the attributes selected by the user in the step from the descriptive attributes to form new target attributes; and repeats the second to fifth steps upon the user's operation.
COPYRIGHT: (C)2004,JPO
摘要翻译：要解决的问题：为了解决尽管引入DNA微阵列能够同时测量许多基因的表达数据的问题，处理大量数据的分析技术是无效的。解决方案：数据分析方法包括从要分析的数据中的多个属性中选择描述属性和目标属性的步骤，分析描述属性与目标属性之间的关系并保存结果的步骤，将分析结果呈现给用户的步骤，从包括在分析结果中的描述性属性中选择用户指定的一个或多个属性的步骤，以及在步骤中去除用户选择的属性的步骤从描述性属性形成新的目标属性; 并在用户操作时重复第二至第五步骤。版权所有（C）2004，JPO

13. 发明专利

JP2010055227A Computer-assisted name identification method and system for the same 审中-公开
标题翻译：计算机辅助名称识别方法和系统
公开(公告)号：JP2010055227A
公开(公告)日：2010-03-11
申请号：JP2008217515
申请日：2008-08-27
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： EGUCHI TOSHIHIRO , ANDO KENJI , MAKI HIDEYUKI
IPC分类号： G06F17/30 , G06Q10/00 , G06Q20/00 , G06Q20/24 , G06Q30/02 , G06Q50/00
摘要： PROBLEM TO BE SOLVED: To achieve high-speed computer-assisted name identification even if the number of customers is enormously increased, and to provide a method for efficiently performing computer-assisted name identification even among multiple customer information DBs. SOLUTION: Provision information of a service provided by a company to a customer is accumulated along with provision time of the service. Settlement information about payment of the value paid by the customer to receive the service is accumulated along with settlement time. History information about service provision is collated with history information of payment, customer information about the service provision is associated with a credit card number used as a payment means, and the credit card number and a customer identification number are associated with the customer information managed by a plurality of companies. In this way, customer identification numbers managed independently by a plurality of companies are associated mutually. COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：为了实现高速计算机辅助名称识别，即使客户数量大大增加，并且提供甚至在多个客户信息DB之间有效地执行计算机辅助名称识别的方法。
解决方案：公司向客户提供的服务的提供信息与服务的提供时间一起累积。关于客户支付接收服务费用的结算信息与结算时间一起累计。关于服务提供的历史信息与付款的历史信息进行整理，关于服务提供的客户信息与用作支付方式的信用卡号相关联，并且信用卡号和客户标识号与由多家公司。以这种方式，由多个公司独立管理的客户识别号码相互关联。版权所有（C）2010，JPO＆INPIT

14. 发明专利

JP2006155074A Access control system 审中-公开
标题翻译：访问控制系统
公开(公告)号：JP2006155074A
公开(公告)日：2006-06-15
申请号：JP2004342750
申请日：2004-11-26
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： MAKI HIDEYUKI , MORITA TOYOHISA , FUKUMOTO YASUSHI , SATO YOSHINORI
IPC分类号： G06F21/24 , G06F12/00 , G06F21/20
摘要： PROBLEM TO BE SOLVED: To limit an access in response to a using purpose of data, even in a user authenticated to access the data. SOLUTION: An access policy is stored in each access policy storage part 14, 24 to specify an access mode for the user with permission given to the using purpose, the user and each data item, according to the using purpose. An access control part 12 of a data server 1 controls the access to a database 13 by a user terminal 2, while being cooperated with an access control part 22 of the user terminal 2. When a user ID of the user is registered in the access policy correlated with the using purpose received from the user via an indication receiving part 23, the use of the data item registered in the database 13 is permitted by the access mode of the data item registered in the access policy while correlated with the using purpose. COPYRIGHT: (C)2006,JPO&NCIPI
摘要翻译：要解决的问题：即使在认证用于访问数据的用户中，也可以响应于数据的使用目的来限制访问。解决方案：访问策略存储在每个访问策略存储部分14,24中，以根据使用目的指定给予用户许可的访问模式，用户和每个数据项。数据服务器1的访问控制部分12在与用户终端2的访问控制部分22协作的同时控制用户终端2对数据库13的访问。当用户的用户ID被登记在访问中时与通过指示接收部分23从用户接收到的使用目的相关联的策略，通过在访问策略中登记的数据项的访问模式来允许使用登记在数据库13中的数据项，同时与使用目的相关联。版权所有（C）2006，JPO＆NCIPI

15. 发明专利

JP2005107621A Data analysis method 审中-公开
标题翻译：数据分析方法
公开(公告)号：JP2005107621A
公开(公告)日：2005-04-21
申请号：JP2003336836
申请日：2003-09-29
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： OKOCHI KAZUYA , MAKI HIDEYUKI , MORITA TOYOHISA
IPC分类号： G06N5/04
摘要： PROBLEM TO BE SOLVED: To efficiently analyze data using a plurality of processors without being affected by devices that process at a slow speed.
SOLUTION: This data analysis method uses an analysis controller and a plurality of analysis execution devices to derive rules that describe the features of data for analysis from the data. The analysis controller performs a partial problem investigation process for dividing a data analysis problem into a plurality of partial problems and calculating the amount of calculations needed to solve each partial problem, and a problem allocation process for calculating an estimated value of the scale of the partial problems that the analysis execution devices can analyze in a certain period of time, on the basis of the amount of calculations needed to solve part of the partial problems and the time needed for the analysis execution devices to solve part of the partial problems, and for determining on the basis of the estimated value the next partial problem to be solved by the analysis execution devices and allocating the next partial problem to be solved to the analysis execution devices. The analysis execution device reports to the analysis controller the amount of calculations for the partial problem solved by the analysis execution devices. The analysis controller performs a process for estimating the amount of problems that the analysis execution devices can analyze in a certain period of time, based on the scale of the partial problems reported from the analysis execution devices.
COPYRIGHT: (C)2005,JPO&NCIPI
摘要翻译：要解决的问题：使用多个处理器有效地分析数据，而不受以慢速处理的设备的影响。解决方案：该数据分析方法使用分析控制器和多个分析执行装置从数据中导出描述用于分析的数据的特征的规则。分析控制器执行部分问题调查处理，用于将数据分析问题划分为多个部分问题并计算解决每个部分问题所需的计算量，以及用于计算部分的等级的估计值的问题分配处理分析执行装置可以在一定时间内分析的问题，根据解决部分问题所需的计算量和分析执行装置解决部分问题所需的时间，以及基于估计值确定由分析执行装置解决的下一个部分问题，并将待解决的下一个部分问题分配给分析执行装置。分析执行装置向分析控制器报告由分析执行装置解决的部分问题的计算量。分析控制器基于从分析执行装置报告的部分问题的规模，执行用于估计分析执行装置可以在一定时间段内分析的问题的量的处理。版权所有（C）2005，JPO＆NCIPI

16. 发明专利

JP2004326641A Feature rule finding method by width priority search 审中-公开
公开(公告)号：JP2004326641A
公开(公告)日：2004-11-18
申请号：JP2003123146
申请日：2003-04-28
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： MAKI HIDEYUKI , OKOCHI KAZUYA , MORITA TOYOHISA
IPC分类号： G06F17/30 , G06N5/04
摘要： PROBLEM TO BE SOLVED: To solve the problem that a redundant rule having an assessment value lower than that of a parent rule can not be excluded in searching a rule having a plurality of condition sections with depth priority. SOLUTION: Research is made with width priority in the order from the rule having the low number of condition sections. The employed rule is stored on an employment rule list. In assessing the adoption or rejection of individual rules, the parent rule of the rule under assessment is searched from the employment rule list at the moment of assessing the adoption or rejection of the individual rules. When the parent rule having the assessment value exceeding that of the rule under assessment is present, the rule under the assessment is not employed. COPYRIGHT: (C)2005,JPO&NCIPI

17. 发明专利

JP2001273315A METHOD FOR DISPLAYING DATA CHARACTERISTIC 失效
公开(公告)号：JP2001273315A
公开(公告)日：2001-10-05
申请号：JP2000087067
申请日：2000-03-23
申请人： HITACHI LTD
发明人： MAKI HIDEYUKI , ITO YUKIYASU
IPC分类号： G06F17/30
摘要： PROBLEM TO BE SOLVED: To solve the problem that it is necessary to effectively support the data analyzing work of a user through mutual communication since it is important to complementarily utilize two kinds of methods of 'hypothesis discovery' and 'hypothesis verification', and it is not sufficient to independently operate each method in data analysis with large amounts of multi-dimensional data as objects. SOLUTION: Data to be used by both data mining as a hypothesis discovering method and an OLAP as a hypothesis verifying method are prepared based on common data definition information. Also, at the time of displaying the data mining result, when a user selects one of the data mining results, the OLAP processed result corresponding to the result is displayed.

18. 发明专利

JP2000293414A METHOD FOR INTEGRATING DIFFERENT KINDS OF DATA SOURCES 失效
公开(公告)号：JP2000293414A
公开(公告)日：2000-10-20
申请号：JP9972799
申请日：1999-04-07
申请人： HITACHI LTD
发明人： MAKI HIDEYUKI , MORITA TOYOHISA , TSUJI HIROSHI , ITO YUKIYASU
IPC分类号： G06F12/00 , G06F17/30
摘要： PROBLEM TO BE SOLVED: To easily conduct arithmetic processing and merge data obtained from different kinds of data sources by merging pierces of table-form data outputted from a data output device and outputting the result as a single piece of table-form data. SOLUTION: There are a plurality of data output processes 101, which input data from one or a plurality of data sources 103 and output table-form data 104 consisting of rows and columns. A data merging process 102 inputs table- form data outputted by a plurality of data output processes 101, and merges and outputs them as a single piece of table-form data 105. The method of designation and merge of the data output processes 101 which are the objects of merge by a data merging process 102 is indicated by a merge setting process 106. Then the data output processes 101 input data from data sources, operate and process them as predetermined and output the results, and the data merging process 102 inputs and merges the data outputted from the data output processes and outputs the result.

19. 发明专利

JPH10312289A METHOD AND DEVICE FOR EXTRACTING/VISUALIZING STRUCTURE RELATED TO RULE 失效
公开(公告)号：JPH10312289A
公开(公告)日：1998-11-24
申请号：JP12048697
申请日：1997-05-12
申请人： HITACHI LTD
发明人： AYUKAWA ERIKA , MAKI HIDEYUKI , MORITA TOYOHISA , MAEDA AKIRA , ITO YUKIYASU
IPC分类号： G06F9/44 , G06N5/04
摘要： PROBLEM TO BE SOLVED: To analyze related structure between variables, related structure between attribute values and related structure between the variable and the attribute value by extracting and visualizing a correlation rule between the variables with attributes by means of providing two types of independent display attributes showing the variables and the attribute values for the apexes of a digraph. SOLUTION: Values inputted from a lower limit evaluation measure input part 206 and a data ID input part 207 are stored in the lower limit evaluation measure storage part 227 of a rule related structure extraction parameter storage device 226 and a data ID storage device 228. Extracted results by an attribute value display-type If-Then rule related structure extraction device 211 are stored in the item storage part 230, the rule storage part 231 and the record storage part 232 of a grouping data storage device 229. A digraph plotting device 212 outputs a pen drawing to an output device 214 by using data stored in the grouping data storage device 229. Thus, the correlation rule with the attribute value can be extracted and visualized.

20. 发明专利

JPH05342255A 失效
公开(公告)号：JPH05342255A
公开(公告)日：1993-12-24
申请号：JP17737892
申请日：1992-06-11
申请人： HITACHI LTD
发明人： MAKI HIDEYUKI , MATSUBA IKUO
IPC分类号： G06F15/18 , G06F17/27 , G06F17/30 , G06N3/00 , G06N99/00 , G06F15/38 , G06F15/40
摘要： PURPOSE:To reduce the skip of retrieving owing to different inscription and a synonym and a retrieving noise owing to the accident of gathering specified character strings in full text research by a natural language. CONSTITUTION:A syntax analytic device 107 analyzes a retrieving condition character string by the syntax analytic device and outputs syntax information (the appearance likelihood of a word and a inter-word distance) in the character string. A goodness- of-fit calcuration device 108 determines a retrieving condition character string inter- word proximate degree based on syntax information in the retrieving condition character string through the use of combination weight in a combination weight file 106, a retrieving objective sentence inter-word proximate degree based on syntax information of a retrieving objective sentence in a document data base 104 through the use of the combination weight, an associative inter-word proximate degree based on the retrieving objective sentence inter-word proximate degree and inter-word associative intensity in a synonym dictionary file 105, a word pair goodness-of-fit based on the retrieving condition character string inter-word proximate degree and the associative inter-word proximate degree through the use of the combination weight and the goodness-of-fit of the retrieving objective sentence by integrating the determined word pair fitness-of-degrees of respective word pairs.

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