-
公开(公告)号：KR1020090012467A
公开(公告)日：2009-02-04
申请号：KR1020070076312
申请日：2007-07-30
Applicant: 한국과학기술정보연구원
IPC: G06F17/30
CPC classification number: G06F17/30887
Abstract: A integrated search system using an URI(Uniform Resource ID) database and a method thereof are provided to output a search result specialized to each entity by storing link information of each entity in an ID system, checking an entity type when a user query is related to the stored entity, and forming a page for each entity according to the entity type. A URI database(40) stores link information corresponding to an entity in an ID system. An entity information processor(10) checks an entity type and provides entity information by searching the URI database according to a query received from a user. An integrated search result provider(30) provides an integrated search result page for the entity based on entity information received from the entity information processor. A mesh-up processor(20) calls an open API(Application Program Interface) based on the entity information received from the entity information processor and performs mesh-up by calling an inference system and using the result.
Abstract translation: 提供了使用URI（统一资源ID）数据库及其方法的集成搜索系统，通过将每个实体的链接信息存储在ID系统中来输出专用于每个实体的搜索结果，当用户查询相关时检查实体类型 到存储的实体，并且根据实体类型为每个实体形成页面。 URI数据库（40）存储与ID系统中的实体相对应的链接信息。 实体信息处理器（10）通过根据从用户接收的查询来搜索URI数据库来检查实体类型并提供实体信息。 综合搜索结果提供者（30）根据从实体信息处理器接收到的实体信息为实体提供综合的搜索结果页面。 网格化处理器（20）基于从实体信息处理器接收到的实体信息来调用开放API（应用程序接口），并通过调用推理系统并使用结果执行网格化。
-
公开(公告)号：KR1020080099895A
公开(公告)日：2008-11-14
申请号：KR1020070045754
申请日：2007-05-11
Applicant: 한국과학기술정보연구원
CPC classification number: G06F17/30005
Abstract: An RDF triple-based class-attribute management system and a method thereof are provided to make a quick and efficient inference service possible. An RDF triple-based class-attribute management system comprises the followings: a class-attribute defining unit(10) which defines the attributes related to each class by using an ontology schema and instances; a class-attribute application part(20) which produces the storage structure suitable for a DBMS with respect to the defined class- attribute relation and generates/stores the relation information; and a class-attribute utilizing part(30) which converts/processes a user query with respect to the relationship information generated in the class-attribute application part by referring to a class-attribution table, and then produces/provides the result suitable for an inference service.
Abstract translation: 提供了一种基于RDF的三类类属性管理系统及其方法，可以实现快速有效的推理服务。 一种基于RDF的三类类属性管理系统包括：一个类属性定义单元（10），通过使用本体模式和实例来定义与每个类相关的属性; 产生/存储关系信息的类属性应用部分（20），其产生适合于针对所定义的类属性关系的DBMS的存储结构; 以及类属性利用部分（30），其通过参考类属性表来转换/处理关于在类属性应用部分中生成的关系信息的用户查询，然后产生/提供适合于 推理服务。
-
公开(公告)号：KR1020080071213A
公开(公告)日：2008-08-04
申请号：KR1020060130706
申请日：2006-12-20
Applicant: 한국과학기술정보연구원
CPC classification number: G06F17/2785 , G06F17/2845
Abstract: A system and a method for providing a research information service based on the semantic web are provided to enable a user to efficiently check all information needed for selecting an optimal coordinates researcher and fixing a correct R&D(Research and Development) policy as well as research trend by constructing an R&D support service platform based on the semantic web. A knowledge processor(100) processes and provides a knowledge information sharing service by processing knowledge based on a URI(Uniform Resource Identifier) and a DBMS(DataBase Management System). An information circulation service unit(200) receives a result from the knowledge processor, and provides registration/search/statistics services of research information based on the semantic web. An inference service unit(300) receives the result from the information circulation service unit, and explores/provides indirect information through fact, rule, and post process. A common service unit includes a URI registration/management service unit registering/managing/searching the URI of manpower, institutions, reports, papers, patents, themes, and categories, and a theme and category classification service unit performing thesaurus base automatic classification.
Abstract translation: 提供了一种基于语义网提供研究信息服务的系统和方法，使用户能够有效地检查选择最佳坐标研究人员所需的所有信息，并修正正确的研发（研发）政策以及研究趋势 通过构建基于语义网的研发支持服务平台。 知识处理器（100）通过基于URI（统一资源标识符）和DBMS（数据库管理系统）处理知识来处理和提供知识信息共享服务。 信息流通服务单元（200）从知识处理器接收结果，并且基于语义网提供研究信息的注册/搜索/统计服务。 推理服务单元（300）从信息流通服务单元接收结果，并通过事实，规则和后期处理来探索/提供间接信息。 公共服务单元包括登记/管理/搜索人力，机构，报告，论文，专利，主题和类别的URI的URI注册/管理服务单元，以及执行同义词库基本自动分类的主题和类别分类服务单元。
-
公开(公告)号：KR1020080049239A
公开(公告)日：2008-06-04
申请号：KR1020060119600
申请日：2006-11-30
Applicant: 한국과학기술정보연구원
CPC classification number: G06F17/30722 , G06F17/30011 , G06N5/003
Abstract: A method for removing disambiguation of homonymic authors by extracting information from original text is provided to remove disambiguation of the homonymic authors by extracting disambiguation solving information from original text of papers of the homonymic authors automatically and utilizing the disambiguation solving information and bibliographic information of the papers. Bibliographic information of papers is automatically extracted from more than one of an author name, a title, a journal name, and an issuance year of the papers when an enormous quantity of scientific and technologic documents is searched(ST1). Disambiguation solving information is automatically from original text of papers of homonymic authors when an enormous quantity of scientific and technologic documents is searched(ST2). Disambiguation among the homonymic authors is removed by using at least one of the bibliographic information and the disambiguation solving information(ST3). The disambiguation solving information is automatically extracted from an email address and institution information of the author, keywords of the paper, a project number, a style of the author, and a terminology set of the original text of the paper.
Abstract translation: 提供了从原文中提取信息来消除同义词作者消歧的一种方法，通过从同声名词作者的原文中提取消歧解决信息来消除同义词作者的歧义，并利用消歧解决信息和论文的书目信息 。 文件的书目信息在搜索大量科学和技术文件（ST1）时，自动从作者姓名，作品名称，日记名称和论文发行年份中提取出来。 消除歧义的解决信息是自动搜索大量科学技术文献（ST2）时同声词作者论文的原文。 通过使用至少一个书目信息和消歧解决信息（ST3）来消除同义词作者之间的歧义。 消歧解决信息是从作者的电子邮件地址和机构信息，论文的关键字，作品的项目编号，作者风格以及论文原文的术语集中自动提取的。
-
公开(公告)号：KR1020080019439A
公开(公告)日：2008-03-04
申请号：KR1020060081785
申请日：2006-08-28
Applicant: 한국과학기술정보연구원
CPC classification number: G06F17/30424 , G06N5/02
Abstract: A system and a method for providing a knowledge extension/inference service based on a DBMS(DataBase Management System) are provided to commercialize a semantic web service by providing a stable and efficient framework, which generates knowledge with ontology, extends the knowledge through an extension rule, and provides an inference service by processing a query. A knowledge generator(10) generates knowledge represented in a triple(Subject-Predicate-Object) by using an ontology schema and instances. A knowledge extender(20) extends the generated knowledge by using a knowledge extension rule. An inference service provider(30) provides an inference service by using the extended knowledge, and converting and processing a user query. The knowledge generator includes an RDF(Resource Description Framework)/XML(eXtensible Markup Language) parser(11) outputting a parsing result by parsing the ontology schema into RDF/XML, a schema-tripe converter(12) performing schema-triple conversion of the parsing result, a DB-triple conversion rule generator(13) generating a DB-triple conversion rule by receiving the schema-triple and DB-class/NS(name space)-prefix mapping information, and a DB-triple converter(14). The DB-triple converter converts ontology instances into ontology instances stored in a URL(Uniform Resource Locator) server into instance triples by using the DB-triple conversion rule.
Abstract translation: 提供了一种用于提供基于DBMS（DataBase Management System）的知识扩展/推理服务的系统和方法，以通过提供一种稳定而有效的框架来商业化语义Web服务，该框架通过本体生成知识，通过扩展来扩展知识 规则，并通过处理查询来提供推理服务。 知识生成器（10）通过使用本体模式和实例生成三重（Subject-Predicate-Object）中表示的知识。 知识扩展器（20）通过使用知识扩展规则来扩展生成的知识。 推理服务提供商（30）通过使用扩展知识提供推理服务，并转换和处理用户查询。 知识生成器包括通过将本体模式解析为RDF / XML来输出解析结果的RDF（资源描述框架）/ XML（可扩展标记语言）解析器（11），执行模式三重转换的模式转换器 解析结果，DB-triple转换规则生成器（13），通过接收模式三和DB-类/ NS（名称空间） - 映射信息生成DB-三重转换规则，以及DB-三重转换器（14） ）。 DB-triple转换器通过使用DB-triple转换规则将本体实例转换为存储在URL（统一资源定位器）服务器中的本体实例到实例三元组。
-
Patent Agency Ranking
公开(公告)号：KR101878854B1
公开(公告)日：2018-07-16
申请号：KR1020170182677
申请日：2017-12-28
Applicant: 한국과학기술정보연구원
CPC classification number: G01N33/0075 , B60Q1/2611 , B60R16/03 , G01N33/0037 , G01N33/0039 , G01N33/004 , G01N33/0042 , G01N33/0047 , G01N33/0062
Abstract: 이동식환경요소측정장치가개시된다. 본발명에따른이동식환경요소측정장치는, 도로를이동하는대중교통수단의외부식별을위해설치되는갓등; 상기갓등의내부공간으로대기가유출입되도록, 상기갓등의전방및 후방에각각관통형성된공기유입공및 공기배출공을포함하는공기순환부; 상기내부공간에탑재된회로기판과, 이에탈착가능하게실장되어상기공기유입공으로유입된대기와관련된환경요소를측정하는측정센서를포함하는센싱모듈; 및상기회로기판에실장되어측정된환경요소정보를대중교통수단의단말장치로송출하고, 대중교통수단의상용전원과연결되어상기센싱모듈에제어전원을인가하는제어부를포함하는것을특징으로한다. 본발명에의하면, 차량의외부식별용갓등과최적화되고컴팩트한센싱모듈이하나의형태로통합되어이동식환경요소장치가구현됨에따라기존에운행중인차량등에간편하게설치및 운용될수 있음은물론이고, 환경요소장치의교체및 수리가용이해지는효과가있으며, 이로인해환경요소정보를보다광범위하고다각적으로수집분석할수 있는기술적토대가마련될수 있고, 나아가생활환경의개선이도모되는효과가있다.
-
公开(公告)号：KR101844914B1
公开(公告)日：2018-04-03
申请号：KR1020170166774
申请日：2017-12-06
Applicant: 한국과학기술정보연구원
Abstract: 본발명은, 초대용량데이터를분산학습하는딥러닝환경에최적화된데이터분배및 셔플링(Shuffling)이수행되도록하는, 새로운딥러닝분산학습을위한데이터관리방안을실현하는기술을개시한다.
-
68.发明授权어댑티브 온톨로지 정보 제공 방법, 어댑티브 온톨로지 정보 제공 장치, 어댑티브 온톨로지 정보를 제공하는 프로그램을 저장하는 저장매체 有权提供自适应本体信息的方法，用于提供自适应本体信息的装置，以及用于存储提供自适应本体信息的程序的存储介质
公开(公告)号：KR101441529B1
公开(公告)日：2014-09-18
申请号：KR1020130020755
申请日：2013-02-26
Applicant: 한국과학기술정보연구원
Abstract: 본 발명은, 어댑티브 온톨로지 정보 제공 방법, 어댑티브 온톨로지 정보 제공 장치, 어댑티브 온톨로지 정보를 제공하는 프로그램을 저장하는 저장매체에 관한 것이다. 본 발명의 일 실시예는, 온톨로지 정보를 저장한 데이터베이스로부터 사용자가 선택한 온톨로지 정보와 관련된 컨택스트 정보를 수신하여 저장하는 단계; 및 상기 저장된 컨택스트 정보에 기반하여, 상기 선택한 M개의 온톨로지 정보들과 관련된 컨택스트 정보들 중 N번 이상 발생한 컨택스트 정보를 어댑티브 온톨로지 정보(adaptive ontology information)로 결정하고, 상기 어댑티브 온톨로지 정보를 제공하는 단계;를 포함하는, 어댑티브 온톨로지 정보 제공 방법을 제공한다.
-
公开(公告)号：KR101396131B1
公开(公告)日：2014-05-19
申请号：KR1020130140217
申请日：2013-11-18
Applicant: 한국과학기술정보연구원
CPC classification number: G06F17/271 , G06F17/2247
Abstract: The present invention relates to an apparatus and a method for measuring the similarity of relation based on a pattern. The apparatus includes a language analysis unit which analyzes the construction, part-of-speech information, and base phrase information of an inputted sentence; a first construction tree generating unit which extracts the construction relation of the sentence based on the results analyzed by the language analysis unit, and generates a first construction tree based on the extracted construction relation; a second construction tree generating unit which reconstructs the first construction tree generated by the first construction tree generating unit into a second construction tree based on a dependency pattern; and a similarity measuring unit which measures similarity by using meaning information and relation information between words existing at an end node of the second construction tree generated by the second construction tree generating unit.
Abstract translation: 本发明涉及一种用于测量基于图案的关系相似度的装置和方法。 该装置包括语言分析单元，其分析输入的句子的构造，部分语音信息和基本信息; 第一构造树生成单元，其基于由语言分析单元分析的结果来提取句子的构造关系，并且基于所提取的构造关系生成第一构造树; 第二构造树生成单元，其基于依赖模式将由所述第一构造树生成单元生成的所述第一构造树重建为第二构造树; 以及相似性测量单元，其通过使用由第二结构树生成单元生成的第二结构树的端节点处存在的词之间的含义信息和关系信息来测量相似度。
-
公开(公告)号：KR101379698B1
公开(公告)日：2014-04-04
申请号：KR1020130072961
申请日：2013-06-25
Applicant: 한국과학기술정보연구원
IPC: A61B5/103
Abstract: The present invention relates to a method and an apparatus for producing the three-dimensional model and physical property information of a human bone, which is capable of photographing human bones to acquire three-dimensional images, forming parts consisting of each bone in the human bones into components, dividing the human bones into eight groups, and performing an indentation test and a compression test on the human bones such as a spine, femoral bones, hip bones, and cranial bones, thereby producing skeletal physical property information. According to the present invention, a method for producing the three-dimensional model and physical property information of a human bone in a system including a video apparatus, a measurement apparatus, and a production apparatus comprises the steps of: (a) allowing the video apparatus to photograph human bones to acquire three-dimensional images; (b) allowing the production apparatus to divide the human bones into five areas of a head part, a spine part, a rib part, an upper limb part, and a lower limb part to classify into cancellous bones and cortical bones, requesting, to the measurement apparatus, an indentation test and a compression test on actual human bones corresponding to the divided areas; (c) allowing the measurement apparatus to perform the indentation test on cortical bones of a spine, femoral bones, hip bones, and cranial bones and the compression test on cancellous bones of the spine and the femoral bones, among the actual human bones corresponding to the divided five areas; and allowing the production apparatus to match the results of the indentation test and the results of the compression test to corresponding human bone images among the three-dimensional images of the human bones and to store the matching information. [Reference numerals] (AA) Start; (BB) End; (S210) Allow an image acquisition unit to photograph human bones to acquire three-dimensional images; (S220) Allow a DB production unit to divide the human bones into eight groups and store the divided information in a skeleton information DB; (S230) Allow a control unit to classify cancellous bones and cortical bones of five areas. relative to three-dimensional images of human bones, and order a skeleton measurement unit to perform an indentation test and a compression test on actual human bones corresponding to the divided areas; (S240) Allow the skeleton measurement unit to perform the indentation test on cortical bones of a spine, femoral bones, hip bones, and cranial bones and the compression test on cancellous bones of the spine and the femoral bones, among the actual human bones corresponding to the divided five areas; (S250) Allow the control unit to match the results of the indentation test and the results of the compression test to corresponding human bone images among the three-dimensional images of the human bones through the DB production unit and to store the matching information in the skeleton information DB
Abstract translation: 本发明涉及一种用于生产人骨三维模型和物理属性信息的方法和装置，其能够拍摄人骨以获取三维图像，在人骨中形成由每个骨组成的部分 将人骨分为8组，对脊骨，股骨，髋骨和颅骨等骨骼进行压痕试验和压缩试验，生成骨骼物理性质信息。 根据本发明，在包括视频设备，测量设备和制作设备的系统中，生成人骨三维模型和物理属性信息的方法包括以下步骤：（a）允许视频 拍摄人骨头以获取三维图像的装置; （b）允许生产设备将人骨分成头部，脊部，肋部，上肢部和下肢部分的五个区域，以分类为松质骨和皮质骨，请求， 测量装置，对应于分割区域的实际人骨的压痕试验和压缩试验; （c）允许测量装置对脊柱，股骨，髋骨和颅骨的皮质骨进行压痕试验，并对脊柱和股骨的松质骨进行压缩试验，对应于实际的人骨骼 分五区; 并且允许生产设备将压痕测试的结果和压缩测试的结果与人骨骼的三维图像中的相应的人骨图像相匹配并存储匹配信息。 （附图标记）（AA）开始; （BB）结束; （S210）允许图像获取单元拍摄人骨以获取三维图像; （S220）允许数据库生成单元将人骨划分为八组，并将分割的信息存储在骨架信息DB中; （S230）允许控制单位对五个区域的松质骨和皮质骨进行分类。 相对于人骨的三维图像，并且命令骨骼测量单元对与分割区域相对应的实际人骨进行压痕测试和压缩测试; （S240）允许骨骼测量单位对脊柱，股骨，髋骨和颅骨的皮质骨进行压痕试验，对脊柱和股骨骨质骨骨折进行压缩试验，实际人骨对应 分五区; （S250）使控制单元通过DB生成单元将压缩测试的结果与压缩测试的结果与人骨的三维图像中的相应的人骨图像相匹配，并将匹配信息存储在 骨架信息DB
-
-
-
-
-
-
-
-
-
Search Results
146
records
(took 0.032 seconds)
