公开(公告)号：JPH08190564A

公开(公告)日：1996-07-23

申请号：JP295595

申请日：1995-01-12

Applicant: IBM JAPAN

Inventor： NOMIYAMA HIROSHI

IPC: G06F17/30

Abstract: PURPOSE: To automatically display what character a set of documents narrowed down at present has by obtaining the number of documents to which keywords are given with respect to a set of keywords given to the set of narrowed-down documents and displaying them in the frequency order. CONSTITUTION: The keyword inputted by a user input and display module 210 is transferred to a keyword retrieval engine 208 to retrieve an ID index 204 to the keyword, and a set of IDs of pertinent documents is returned. The keyword retrieval result (the number of retrieved documents, titles, or the like) is displayed in individual windows by the module 210. Obtained keywords are sorted in the descending order based on counted values by a keyword collection and sort module 212 and are displayed by the module 210. A list of keywords sorted in the descending order in this manner is watched to find what character the set of narrowed-down documents has.

公开(公告)号：JP2009129296A

公开(公告)日：2009-06-11

申请号：JP2007305280

申请日：2007-11-27

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： SUZUKI AKIKO ,  NOMIYAMA HIROSHI ,  SATO ATSUSHI ,  HAYASHIDA NORIMASA

IPC: G06F17/30 ,  G06F21/24

CPC classification number: G06Q30/02 ,  G06Q10/10 ,  G06Q50/265

Abstract: PROBLEM TO BE SOLVED: To enable a user to set whether or not to disclose preference information of the user depending on preference information of other users or a user group. SOLUTION: By applying a privacy policy, which describes and manages a policy for disclosure of preference information, to the preference information, it is determined whether or not to disclose some or all of the information to a third party based on classification as a result of matching by a matching system. The method includes: disclosing, to a user who has made a query, only preference information that matches preferences of the user; and disclosing only preference information common to the user group. The description of a privacy policy makes it possible to achieve a meaningful communication by disclosing preference information to others within a detailed range while keeping the preference information undisclosed if the user does not wish disclosure thereof, such as disclosing only preference information including preferences matching those of the other user or disclosing only preference information including preferences shared in a certain user group. COPYRIGHT: (C)2009,JPO&INPIT

Abstract translation: 要解决的问题：为了使用户能够根据其他用户或用户组的偏好信息来设置是否公开用户的偏好信息。

解决方案：通过应用隐私策略，其描述和管理偏好信息的公开的策略，对偏好信息，确定是否根据分类将其部分或全部信息公开给第三方 匹配系统匹配的结果。 该方法包括：向已经进行查询的用户披露与用户的偏好匹配的偏好信息; 并且仅公开用户组通用的偏好信息。 隐私政策的描述使得有可能通过在详细范围内向其他人公开偏好信息来实现有意义的通信，同时如果用户不希望公开那些保持优先信息不被公开的优先信息，例如仅披露包括与 另一个用户或仅披露包括在某个用户组中共享的偏好的偏好信息。 版权所有（C）2009，JPO＆INPIT

公开(公告)号：JP2007102289A

公开(公告)日：2007-04-19

申请号：JP2005287701

申请日：2005-09-30

Applicant: Ibm Japan Ltd ,  日本アイ・ビー・エム株式会社

Inventor： TSUTSUMI TAIJIRO ,  TAKEDA KOICHI ,  NOMIYAMA HIROSHI

IPC: G06F17/30

CPC classification number: G06Q40/06

Abstract: PROBLEM TO BE SOLVED: To precisely retrieve text information which has strong association with numerical information such as time series data, and to display the text information by associating it with the numerical information. SOLUTION: This information processor for associating text information with the numerical information of time series is provided with a generation part for generating first words and phrases showing the change of a first numerical information in a certain period; a first retrieval part for retrieving text information matched with a retrieval key including the first words and phrases from among a plurality of pieces of text information as the object of retrieval; and an output part for outputting the retrieved text information by associating it with the first numerical information in the period. In this information processor, the first retrieval part may retrieve text information describing the fact that the first numerical information has been changed as shown by the first words and phrases from among a plurality of pieces of text information prepared or announced in the period. COPYRIGHT: (C)2007,JPO&INPIT

Abstract translation: 要解决的问题：精确检索与诸如时间序列数据的数字信息强关联的文本信息，并通过将文本信息与数字信息相关联来显示文本信息。 解决方案：用于将文本信息与时间序列的数字信息相关联的信息处理器具有生成部件，用于生成在一定时间段内显示第一数字信息的改变的第一单词和短语; 第一检索部分，用于检索与包括作为检索对象的多条文本信息之中的第一单词和短语的检索关键字匹配的文本信息; 以及输出部，其通过将检索到的文本信息与该期间中的第一数值信息相关联来输出。 在该信息处理器中，第一检索部分可以检索描述第一数字信息已被改变的文本信息，如在该期间准备或宣布的多条文本信息中的第一个单词和短语所示。 版权所有（C）2007，JPO＆INPIT

公开(公告)号：JP2002230035A

公开(公告)日：2002-08-16

申请号：JP2001000815

申请日：2001-01-05

Applicant: IBM

Inventor： NOMIYAMA HIROSHI ,  TAKEDA KOICHI ,  SAKAIRI TAKASHI

IPC: G06F17/30 ,  G06F7/00

Abstract: PROBLEM TO BE SOLVED: To present valuable information to a user at the time of periodically observing a plurality of dynamically changing information sources. SOLUTION: This information processing system for processing information to be acquired from a plurality of sites connected through the Internet 10 is provided with a crawler 13 patrolling in the sites on the Internet 10 registered in a registered site DB11, a meta data DB12 for storing meta data to be acquired according to the patrol by extracting elements carrying information from contents to be referred to by a URL, a significant information element extracting mechanism 30 for reading the information stored in the meta data DB12, and for extracting significant information elements based on the level of coincidence of the information elements, a significant information element DB40 for storing the extracted significant information, and a result display mechanism 41 for operating visualization processing to the stored significant information elements.

公开(公告)号：JP2002197118A

公开(公告)日：2002-07-12

申请号：JP2000382588

申请日：2000-12-15

Applicant: IBM

Inventor： ASAKAWA CHIEKO ,  TAKAGI HIRONOBU ,  NOMIYAMA HIROSHI

IPC: G06F17/30 ,  G06F3/16 ,  G06F17/22 ,  G09G5/00

Abstract: PROBLEM TO BE SOLVED: To effectively obtain information and provide an effective means for presenting the obtained information without a definite indication of an intention for obtaining the information, such as pushing down of a retrieval button. SOLUTION: A kana (Japanese syllabary)/kanji (Chinese character) converting routine is started (step 50), and a character string is inputted (step 51). Next, a conversion key for converting the inputted character string to kanji is pushed down (step 52). In this timing for pushing down the conversion key, a homonym candidate selection routine is started (step 53), and a conversion candidate is presented. In the timing for pushing down the conversion key or conversion operation of the conversion candidate in the homonym candidate selection routine, namely candidate pre-selection operation (step 54), an information access routine is started (step 55). Next, information access is executed (step 56), and the retrieval result (the information obtained by access) is obtained (step 57). Then, the retrieval result is presented (step 58).

公开(公告)号：JPH04205058A

公开(公告)日：1992-07-27

申请号：JP33098890

申请日：1990-11-30

Applicant: IBM

Inventor： NOMIYAMA HIROSHI

IPC: G06F17/27

Abstract: PURPOSE: To enable the execution of machine translation with which the merit of CBMT is provided and processing efficiency is improved by generalizing knowledge provided from translated examples and using such generalized knowledge for translation processing. CONSTITUTION: In order to find the high-order conception of a word, first of all, it is investigated whether any translation pattern and thesaurus having the same number as its translation pattern description are existent or not and when they are existent, that translation pattern thesaurus is retrieved before the thesaurus. Thus, by searching the generalized translation knowledge, shortest distance generalized translation knowledge is provided. In the case of retrieving the shortest distance generalized knowledge to be applied to inputting, it is not necessary to calculate the degree of similarity.

公开(公告)号：JPH0447364A

公开(公告)日：1992-02-17

申请号：JP15557090

申请日：1990-06-15

Applicant: IBM

Inventor： NAGAO KATASHI ,  NOMIYAMA HIROSHI

IPC: G06F17/27 ,  G06F17/28

Abstract: PURPOSE: To construct knowledge for canceling the structural equivocality of natural language by extracting pairs of modifying word and modified words concerning respective plural possible dependence relations concerning a sentence, for which structural equivocality is discriminated, and deciding the dependence relation of the highest likelihood based on the distance found for each pair. CONSTITUTION: The structural equivocality to be the maximum bottle neck for analyzing the natural language sentence is generated because plural modification (dependence) relations between vocabularies can be considered. Then, first of all, the knowledge is expressed in three structure showing synonymous relation, hierarchical relation and dependence relation. Concerning the equivocal modification, the dependence relation between vocaburaries defined in the background knowledge is searched by a dependence relation analyzer while using the synonymous relation or hierarchical relation, the optimum modification is selected while using limitations provided from the sentence and limitations provided from contexts, and equivocality is canceled. The fixed dependence structure is registered in a knowledge base as context dependence relation data. Thus, the knowledge for canceling structural equivocality in natural language can be constructed.

公开(公告)号：DE69627058D1

公开(公告)日：2003-05-08

申请号：DE69627058

申请日：1996-01-10

Applicant: IBM

Inventor： NOMIYAMA HIROSHI

IPC: G06F17/30

Abstract: The present invention provides, in a database, a technique for automatically indicating what property a current set of documents narrowed by a search criteria possesses, and an approach for retrieving a set of similar documents by specifying a document rather than a keyword. First, in a set of keywords attached to the set of documents narrowed in response to the user retrieval processing, a system according to preferred embodiments of the present invention counts the number of documents to which each of the keywords is attached and displays the keywords in order of decreasing frequency. Next, a user specifies his/her document(s) of interest among the set of documents. In response to this, out of the keywords attached to the documents specified by the user, a keyword able to retrieve any documents other than are specified is displayed. From the set of keywords attached to the current set of document narrowed, this is determined by omitting the keywords meeting the following conditions: (1) the keyword is not attached to any documents other than the specified one; and (2) the keyword is attached to all the documents in the current set of documents. An automatic retrieval with this keyword will enable narrowing by document rather by the keyword.

公开(公告)号：JP2010211821A

公开(公告)日：2010-09-24

申请号：JP2010111458

申请日：2010-05-13

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： NOMIYAMA HIROSHI ,  TAKUMA DAISUKE

IPC: G06F17/30

Abstract: PROBLEM TO BE SOLVED: To retrieve a document data while appropriately reflecting the content of a retrieving sentence, and to appropriately detect the occurrence of a problem out of document data sequentially added. SOLUTION: This retrieval system retrieves the document data including the content of the retrieving sentence from the plurality of document data, and includes a document database for storing the plurality of document data, a concept database for storing a plurality of concepts by hierarchical structure, a document data concept extraction part for extracting document concept corresponding to document data, based on a keyword included in each document data, a retrieving sentence concept extraction part for extracting a retrieving sentence concept, based on the keyword included in the retrieving sentence, a concept retrieval part for retrieving the document data with the retrieving sentence concept serving as an upper hierarchy or lower hierarchy of the document concept, out of the plurality of document data, and a retrieval result output part for outputting the document data retrieved by the concept retrieval part, as the document data including the content assigned by the retrieving sentence. COPYRIGHT: (C)2010,JPO&INPIT

Abstract translation: 要解决的问题：在适当地反映检索句子的内容的同时检索文档数据，并且从顺序添加的文档数据中适当地检测问题的发生。 解决方案：该检索系统从多个文档数据中检索包括检索句子的内容的文档数据，并且包括用于存储多个文档数据的文档数据库，用于通过分层存储多个概念的概念数据库 基于包括在每个文档数据中的关键词提取与文档数据相对应的文档概念的文档数据概念提取部分，用于提取检索句子概念的检索语句提取部分，基于检索句子中包含的关键字， 一种概念检索部分，用于利用在多个文档数据中作为文档概念的上层或下层的检索句子概念检索文档数据，以及检索结果输出部分，用于输出由概念检索的文档数据 检索部分，作为包括通过检索分配的内容的文档数据 句子。 版权所有（C）2010，JPO＆INPIT

公开(公告)号：JP2006268751A

公开(公告)日：2006-10-05

申请号：JP2005089579

申请日：2005-03-25

Applicant: Ibm Japan Ltd ,  日本アイ・ビー・エム株式会社

Inventor： MARUYAMA HIROSHI ,  ITOI MASAHARU ,  NOMIYAMA HIROSHI

IPC: G06F21/24 ,  G06F17/30

CPC classification number: G06F21/6263

Abstract: PROBLEM TO BE SOLVED: To present information regarding a file that has been secondly and derivatively created from a leaked file, by retrieving this derivatively created file from the firstly leaked file, and to present personal information recorded in a computer accessible via a broadband network. SOLUTION: The disclosed apparatus comprises: a recording section for recording original data of a leaked file; an extraction section for extracting a representation corresponding to information, to prevent leakage, contained in the original data and files present in a recording region to be investigated; an investigation section for investigating a degree of relation between the files present in the recording region to be investigated and the original data; and a presentation section for presenting information, based on the degree of relation, regarding a file created based on the leaked file. COPYRIGHT: (C)2007,JPO&INPIT

Abstract translation: 要解决的问题：通过从第一泄漏文件中检索该衍生创建的文件来呈现关于从泄密文件二次并衍生创建的文件的信息，并且呈现记录在可通过以下操作访问的计算机中的个人信息： 宽带网络。 解决方案：所公开的装置包括：记录泄漏文件的原始数据的记录部分; 提取部分，用于提取对应于包含在待调查的记录区域中的原始数据和文件中的信息的防止泄漏的表示; 调查部分，用于调查存在于待调查记录区域中的文件与原始数据之间的关系程度; 以及呈现部分，用于基于关于基于所泄漏的文件创建的文件的关系的程度呈现信息。 版权所有（C）2007，JPO＆INPIT