公开(公告)号：JP2013077157A

公开(公告)日：2013-04-25

申请号：JP2011216500

申请日：2011-09-30

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation

Inventor： ISHO TOSHIO ,  ITO TAKASHI ,  OSOGAMI TAKAYUKI

IPC: G06K9/03

CPC classification number: G06K9/03

Abstract: PROBLEM TO BE SOLVED: To provide a system, method and program for supporting proof reading of text data generated through optical character recognition.SOLUTION: A system includes: a first proof reading tool for performing carpet proof reading on text data; a second proof reading tool for performing side-by-side proof reading on the text data; a storage unit configured to store a log of proof reading operations having been performed by using the first and second proof reading tools; and an analysis unit configured to determine, for each attribute serving as units in which carpet proof reading is performed with the first proof reading tool, whether or not to use the first proof reading tool in proof reading of the attribute, by comparing a first estimated value of time taken when proof reading is performed by using the first proof reading tool with a second estimated value of time taken when proof reading is performed by using the second proof reading tool without using the first proof reading tool, the first and second estimated values being calculated on the basis of the log.

Abstract translation: 要解决的问题：提供用于支持通过光学字符识别产生的文本数据的校对读取的系统，方法和程序。 解决方案：一种系统包括：用于对文本数据进行地毯识别读取的第一校验读取工具; 用于对文本数据执行并排验证读取的第二证明阅读工具; 存储单元，被配置为存储已经通过使用所述第一和第二证明读取工具执行的证明读取操作的日志; 以及分析单元，其被配置为，通过比较第一估计值和第一估计值，确定作为用第一证明读取工具执行地毯证明读取的单元的每个属性是否使用第一证明读取工具进行证明读取的属性 通过使用第一校样读取工具执行校样读取时所用的时间的值，其中使用第二校样读取工具执行校样读取时所花费的时间的第二估计值，而不使用第一校样读取工具，第一和第二估计值 根据日志计算。 版权所有（C）2013，JPO＆INPIT

公开(公告)号：JP2007249754A

公开(公告)日：2007-09-27

申请号：JP2006074221

申请日：2006-03-17

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ASAKAWA CHIEKO ,  ITO TAKASHI ,  TAKAGI HIRONOBU ,  ISHIHARA TATSUYA

IPC: G06F3/048 ,  G06F3/16 ,  G06F12/00

CPC classification number: G06F9/4443

Abstract: PROBLEM TO BE SOLVED: To provide a user interface for operating a plurality of application programs in common. SOLUTION: The information processor for providing a common input/output interface to a plurality of application programs includes: a document conversion section for converting a specific document created by each of the plurality of application programs and represented by a data structure specific to the each application program into a common document represented by a common data structure; an output section for outputting the common document to a user; an input section for inputting the user operation for the common document; and an interface adapter section for converting the object included in the common document into the object in the output section; and a modifying section for modifying the common document according to the content of the user operation; and a modification reflecting section for reflecting the modification to the common document in the specific document. COPYRIGHT: (C)2007,JPO&INPIT

Abstract translation: 要解决的问题：提供用于共同操作多个应用程序的用户界面。 解决方案：用于向多个应用程序提供公共输入/输出接口的信息处理器包括：文档转换部分，用于转换由多个应用程序中的每一个创建的特定文档，并由特定于 将每个应用程序转换成由通用数据结构表示的通用文档; 输出部，用于向用户输出公共文档; 用于输入所述公共文档的用户操作的输入部分; 以及接口适配器部分，用于将包括在公共文档中的对象转换成输出部分中的对象; 以及修改部，用于根据用户操作的内容修改公共文档; 以及用于将修改反映到特定文档中的共同文档的修改反映部分。 版权所有（C）2007，JPO＆INPIT

公开(公告)号：JP2003085087A

公开(公告)日：2003-03-20

申请号：JP2001259846

申请日：2001-08-29

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーション

Inventor： TAKAGI HIRONOBU ,  ITO TAKASHI

IPC: G06F13/00 ,  G06F17/30

CPC classification number: G06F17/30905 ,  Y10S707/99937 ,  Y10S707/99945

Abstract: PROBLEM TO BE SOLVED: To effectively reduce manhours required for annotation in transcoding by appropriately applying the annotation to multiple Web contents.
SOLUTION: This transcoding system is provided with an annotation database 20 storing the annotation to be referred for converting the contents and a transcoder 10 converting the contents based on the annotation stored in the annotation database 20. The annotation database 20 selects the annotation applicable to the contents based on the correspondence between the constitution elements of the contents to be processed and description of the annotation according to the request from the transcoder 10 and returns it to the transcoder 10.
COPYRIGHT: (C)2003,JPO

Abstract translation: 要解决的问题：通过适当地将注释应用于多个Web内容，有效地减少转码中注释所需的工作时间。 解决方案：该代码转换系统被提供有存储用于转换内容的要引用的注释的注释数据库20和基于存储在注释数据库20中的注释来转换内容的代码转换器10.注释数据库20选择适用于 基于要处理的内容的构成元素与代码转换器10的请求的注释的描述之间的对应关系的内容，并将其返回到代码转换器10。

公开(公告)号：JP2003058498A

公开(公告)日：2003-02-28

申请号：JP2001233963

申请日：2001-08-01

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーション

Inventor： ITO TAKASHI

IPC: G06F15/00 ,  G06F13/00 ,  H04L29/06 ,  H04L29/08 ,  H04L29/12

CPC classification number: H04L67/34 ,  H04L29/06 ,  H04L29/12009 ,  H04L29/12594 ,  H04L61/30 ,  H04L67/02 ,  H04L67/14 ,  H04L67/142 ,  H04L69/329

Abstract: PROBLEM TO BE SOLVED: To provide a session management method in an environment disabled from utilizing a cookie capable of utilizing static pages, minimizing the reconstruction load of an already constructed web site, reducing server's load, and suppressing the increment of page size of a web page. SOLUTION: A virtual directory name is issued in starting a session and redirected in a page existing after the virtual directory name. During the trace of the session, the virtual directory name is converted into a substantial directory name (step 58), a method is executed in a resource specified by the substantial directory name (step 59). In executing the method, a user accessed by the virtual directory name is specified as necessary (step 60). Movement (i.e., movement in session management) in the substantial directory constituted so as to exist after the fictional directory is carried out by link movement by a relative pass.

Abstract translation: 要解决的问题：为了在使用能够利用静态页面的cookie禁用的环境中提供会话管理方法，使已经构建的网站的重建负载最小化，减少服务器的负载以及抑制网页的页面大小的增加 页。 解决方案：在启动会话并在虚拟目录名称之后存在的页面中重定向时，会发出虚拟目录名称。 在会话的跟踪期间，将虚拟目录名称转换为实质的目录名称（步骤58），在由实质目录名称指定的资源中执行方法（步骤59）。 在执行该方法时，根据需要指定虚拟目录名称访问的用户（步骤60）。 通过相对通过的链接移动来执行在虚构目录之后存在的实质目录中的移动（即，会话管理中的移动）。

公开(公告)号：JP2007072528A

公开(公告)日：2007-03-22

申请号：JP2005255548

申请日：2005-09-02

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ISHIHARA TATSUYA ,  TAKAGI HIRONOBU ,  ITO TAKASHI ,  ASAKAWA CHIEKO

IPC: G06F17/30

CPC classification number: G06F17/30056 ,  G06K9/00463

Abstract: PROBLEM TO BE SOLVED: To improve the accuracy of reading out by voice and text mining by automatically performing structuring of a presentation file.
SOLUTION: The device, control method and program comprise an overlap grouping part extracting overlap information between objects within the presentation file and grouping the information as parent-child relation; a graph division grouping part showing the objects as nodes of a graph, and recursively performing graph division so that a predetermined cost between nodes is minimized, thereby grouping the objects as fraternity; a distance information grouping part further grouping, when distance information between the object is a threshold determined by a predetermined calculation from a distribution histogram of distance information or less, the distance information as fraternity; and a link information extraction part extracting an arrow pattern showing a link relation and generating the link relation and a link label as link information, and outputs resulting structured data as meta information.
COPYRIGHT: (C)2007,JPO&INPIT

Abstract translation: 要解决的问题：通过自动执行演示文件的构造，提高通过语音和文本挖掘读出的准确性。 解决方案：设备，控制方法和程序包括重叠分组部分，提取呈现文件内的对象之间的重叠信息，并将该信息分组为母子关系; 将对象作为图形的节点示出的图形分割分割部分，并且递归地执行图形划分，使得节点之间的预定成本最小化，从而将对象分组为博爱; 距离信息分组部分，当所述对象之间的距离信息是由距离信息的分布直方图或更小的预定计算确定的阈值时，进一步分组，所述距离信息为博爱; 以及链接信息提取部分，提取表示链接关系并生成链接关系的箭头图案和链接标签作为链接信息，并将所得到的结构化数据作为元信息输出。 版权所有（C）2007，JPO＆INPIT

公开(公告)号：JP2004341671A

公开(公告)日：2004-12-02

申请号：JP2003135413

申请日：2003-05-14

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： FUKUDA KENTARO ,  NAKAIKE TAKUYA ,  TAKAGI HIRONOBU ,  ITO TAKASHI

IPC: G06F11/28 ,  G06F17/22 ,  G06F17/27

CPC classification number: G06F17/2247 ,  G06F17/27

Abstract: PROBLEM TO BE SOLVED: To appropriately detect an error part of a program. SOLUTION: An information processing system for analyzing a generation program for generating document information comprising a plurality of document components, which generation program has a plurality of generation commands for generating the plurality of document components included in the document information respectively, comprises: a correspondence information generation part for executing the generation program and generating correspondence information indicating correspondence between the plurality of generation commands and the document components generated by the executed generation commands; a selection part for selecting any of the plurality of document components at a user's instruction; and a command information output part for selecting, according to the correspondence information, the generation command that has generated the document component selected by the selection part and outputting information identifying the generation command. COPYRIGHT: (C)2005,JPO&NCIPI

公开(公告)号：JPH1173298A

公开(公告)日：1999-03-16

申请号：JP23071497

申请日：1997-08-27

Applicant: IBM

Inventor： ITO TAKASHI ,  ASAKAWA CHIEKO

IPC: G06F3/16 ,  G06F17/21 ,  G09B21/00 ,  G10L13/04 ,  G10L5/02

Abstract: PROBLEM TO BE SOLVED: To voice-output text data by sufficiently realizing both the quick reading of the text data and the support of the content capture of the text data of a listener. SOLUTION: A computer 16 outputs a voice signal for reading text data at a normal speed (for example, a speed at which an announcer reads news in radio broadcasting) when a user depresses a normal mode key 180 of a keyboard 18 (a normal mode). Also, the computer 16 converts only a range (audible range) valid for the content understanding of the user in the text data into a voice signal at an audible fast-forwarding speed (which is about 1.5-3 times as fast as a normal speed), and outputs it after the user presses a fast- forwarding mode key 182, and a transition time P passes. Also, the computer 16 converts the other area (a skip area) into a voice signal at a speed (a skip speed which is more than three times as fast as the normal speed) in a level that at least the length and note of the text data can be captured, and outputs it regardless of whether or not the user can listen to it at a high speed (a skip mode).