-
公开(公告)号：KR1020110061229A
公开(公告)日：2011-06-09
申请号：KR1020090117819
申请日：2009-12-01
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2735 , G06F17/278 , G06F17/30731
Abstract: PURPOSE: A system and method for semi-automatic construction of acronym dictionary is provided to improve the language analysis performance by semi-automatically finding an abbreviation word and an original word of the abbreviation word. CONSTITUTION: Some parts of an electronic document are extracted(S210). A proper noun which is included in the electronic document is acknowledged(S220). An abbreviation word about a proper noun candidate is extracted(S230). The abbreviation word candidates are listed in priority order(S240). An abbreviation word is selected(S250). The selected abbreviation word is stored into an abbreviation dictionary database.
Abstract translation: 目的：提供半自动构建首字母缩写词典的系统和方法，通过半自动找到缩写词的缩写词和原始词来提高语言分析性能。 规定：提取电子文档的某些部分（S210）。 电子文档中包含的专有名词被确认（S220）。 提取关于专名候选人的缩写词（S230）。 缩写词候选者以优先顺序列出（S240）。 选择缩写词（S250）。 所选择的缩写词被存储在缩写字典数据库中。
-
公开(公告)号：KR1020110040147A
公开(公告)日：2011-04-20
申请号：KR1020090097306
申请日：2009-10-13
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30654
Abstract: PURPOSE: An apparatus for answering to a question based on answer trustworthiness and a method thereof are provided to evaluate the reliability for the correct answer candidates to a user query in score and reflect the scored reliability to the correct answer priority in order to provide a user with a reliable correct answer. CONSTITUTION: A correct answer indexing unit(20) indexes the documents of which document reliability satisfies a critical value, and stores the indexed documents in a knowledge storage unit. A correct answer candidate extraction unit(40) extracts the correct answer candidate documents for the user query from the knowledge storage unit. A correct answer source reliability measurement unit(53) analyzes a non-textual feature of the correct answer candidate documents in order to measure source reliability. A correct answer extraction strategy reliability measurement unit(55) analyzes the validity of the extraction strategy of the correct answer candidate documents to measure the extraction strategy reliability.
Abstract translation: 目的：提供一种基于应答可信赖性回答问题的方法及其方法，以评估用户查询的正确答案候选人的得分可靠性，并将得分可靠性反映到正确答案优先级，以提供用户 有一个可靠的正确答案。 构成：正确答案索引单元（20）对文件的可靠性满足临界值的文档进行索引，并将索引的文档存储在知识存储单元中。 正确答案候选提取单元（40）从知识存储单元提取用于用户查询的正确答案候选文档。 正确答案源可靠性测量单元（53）分析正确答案候选文件的非文本特征以便测量源可靠性。 正确答案提取策略可靠性测度单元（55）分析正确答案候选文件的提取策略的有效性，以测量提取策略的可靠性。
-
公开(公告)号：KR1020100068964A
公开(公告)日：2010-06-24
申请号：KR1020080127490
申请日：2008-12-15
Applicant: 한국전자통신연구원
IPC: G06F17/30
CPC classification number: G06F17/30867 , G06F17/3053 , G06F17/30536 , G06F17/30887
Abstract: PURPOSE: An apparatus for recommending related query and a method thereof are provided to use a click log of a search engine thereby suggesting related query language correlating with inputted initial query language. CONSTITUTION: An information extraction unit(112) extracts a plurality of each different query languages, URL(Uniform Resource Locator) information selected by the query languages and time information selected each URL reference to a click log. An index unit(116) calculates relation between the URL and the query language and generates click log index data. If the query language is inputted, a server controller(110) searches related URL and query language reference to the click log index data and classifies the related query language by a category and recommends a related query language having correlation with the query language relatively.
Abstract translation: 目的：提供一种用于推荐相关查询的装置及其方法，以使用搜索引擎的点击日志，从而建议与输入的初始查询语言相关的相关查询语言。 构成：信息提取单元（112）提取多个每个不同的查询语言，由查询语言选择的URL（统一资源定位符）信息和选择每个URL引用的时间信息到点击日志。 索引单元（116）计算URL和查询语言之间的关系，并生成点击日志索引数据。 如果输入了查询语言，则服务器控制器（110）搜索相关的URL和查询语言参考点击日志索引数据，并按类别对相关查询语言进行分类，并推荐与查询语言相关的相关查询语言。
-
公开(公告)号：KR1020100066920A
公开(公告)日：2010-06-18
申请号：KR1020080125438
申请日：2008-12-10
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30011 , G06F17/30613 , G06F17/30705
Abstract: PURPOSE: An electronic document processing device and a method thereof are provided to determine duplicated document according to duplicate sentence rate of electronic document and reduce target electronic document effectively, thereby increasing efficiency of query response. CONSTITUTION: A sentence separation block(106) separates each sentence in extracted body content. A duplicated document decision block(108) changes the separated documents through hash algorithm to inherent hash value. According to collision between the changed hash value and pre-stored hash value, the duplicated document decision block determines duplicated sentence. The duplicated document decision block determines duplicated document according to duplicated document ratio of the electronic document.
Abstract translation: 目的：提供电子文件处理装置及其方法，根据电子文件的重复句子率确定重复的文件，有效减少目标电子文件，从而提高查询响应的效率。 构成：句子分离块（106）分离提取的身体内容中的每个句子。 复制文档决策块（108）通过散列算法将分离的文档改变为固有散列值。 根据改变的哈希值和预先存储的哈希值之间的冲突，复制的文档决定块确定重复的句子。 复制的文档决定块根据电子文档的重复文档比例确定重复的文档。
-
公开(公告)号：KR1020080095180A
公开(公告)日：2008-10-28
申请号：KR1020080035896
申请日：2008-04-18
Applicant: 한국전자통신연구원 , 건국대학교 산학협력단
IPC: G06F17/30
CPC classification number: G06F17/30781 , G06F17/30038 , G06F17/30265 , G06F17/30749 , G06F17/30755
Abstract: A method and an apparatus for retrieving multimedia contents are provided to analyze the meaning of an inquiry of a user correctly in a retrieving operation, thereby correctly retrieving multimedia contents corresponding to the inquiry. An inquiry of a user is represented by using a pointer which points a specific region of an MPEG-7 document and a reference which refers to the pointer(10). The meaning of the inquiry represented by using the pointer and the reference is analyzed(20). Multimedia contents corresponding to the inquiry are retrieved according to the analysis result(30).
Abstract translation: 提供一种用于检索多媒体内容的方法和装置，用于在检索操作中正确地分析用户查询的含义，从而正确地检索对应于查询的多媒体内容。 通过使用指向MPEG-7文档的特定区域的指针和引用指针（10）的引用来表示用户的查询。 分析使用指针和参考表示的查询的含义（20）。 根据分析结果（30）检索与查询对应的多媒体内容。
-
Patent Agency Ranking
公开(公告)号：KR100852174B1
公开(公告)日：2008-08-13
申请号：KR1020070014774
申请日：2007-02-13
Applicant: 한국전자통신연구원
IPC: G06F17/30
Abstract: 본 발명은 계층적 분류에 의한 정보 표시 장치에서 정보를 표시하는 방법에 있어서,적어도 2개의 미리 결정된 온톨로지(Ontology) 의미 구조 클래스(Class)에 상응하는 제1 분류 기준이 선택되는 단계, 상기 선택된 각각의 제1 분류 기준에 상응하는 정보를 검색하는 단계, 상기 검색된 정보를 상기 제1 분류 기준의 하위 온톨로지(Ontology) 의미 구조 클래스(Class)인 제2 분류 기준에 상응하여 분류하는 단계, 상기 적어도 2개의 제1 분류 기준을 각각 한 축으로 하고 상기 제2 분류 기준을 축의 성분으로 하는 행렬을 생성하는 단계, 상기 성분이 선택되면 상기 성분에 상응하는 적어도 2개의 제2 분류기준을 각각 한 축으로 하고 상기 제2 분류 기준의 하위 온톨로지(Ontology) 의미 구조 클래스(Class)를 축의 성분으로 하는 하위 행렬을 생성하는 단계를 포함하되, 상기 생성된 행렬의 각 원소는 상기 원소의 위치에 상응하는 적어도 2개의 제2 분류 기준에 의해 동시에 분류된 정보를 포함하고, 상기 행렬의 각 원소는 타일바(Tile Bar) 형태로 구성되며, 상기 타일바는 상기 타일바에 상응하는 원소가 포함하는 정보의 양에 따라 각각 상이한 색상을 가지는 것을 특징으로 하는 계층적 분류에 의한 정보 표시 방법을 제공한다.
정보 표시, 계층, 분류, 온톨로지(Ontology)-
公开(公告)号：KR100831055B1
公开(公告)日：2008-05-20
申请号：KR1020060094538
申请日：2006-09-28
Applicant: 한국전자통신연구원
Abstract: 본 발명은 데이터를 검색할 적어도 하나의 선택할 수 있는 구분자(Facet)를 표시하는 단계 및 적어도 하나의 선택된 구분자(Facet)에 상응하여 데이터를 온톨로지(Ontology) 방식에 의해 검색하고 선택하는 단계를 포함하되, 구분자는 상위 개념부터 하위 개념까지 트리 구조로 표시되는 것을 특징으로 하는 온톨로지 기반의 정보 검색 방법을 제공할 수 있다.
온톨로지(Ontology), 정보 검색-
公开(公告)号：KR1020070098469A
公开(公告)日：2007-10-05
申请号：KR1020070008812
申请日：2007-01-29
Applicant: 한국전자통신연구원
IPC: G06F17/30
CPC classification number: G06F17/30023
Abstract: A device and a method for searching multimedia with metadata are provided to search the multimedia data easily without directly receiving a query made in the MPEG(Moving Picture Experts Group)-7 metadata from a user by using query property of an MPEG-7 query format and mapping information of the MPEG-7 metadata. A mapping information storing part(40) stores/manages the mapping information between the MPEG-7 query property and MPEG-7 metadata items. A query property mapper(20) obtains the MPEG-7 metadata items mapped to the MPEG-y query property according to the user query by using the mapping information. A query input part(10) generates and outputs an MPEG-7 query according to the MPEG-7 query format from the user query. The query property mapper generates and outputs the MPEG-7 metadata query by using the mapped MPEG-7 metadata items. A searcher(30) searches the multimedia by using the MPEG-7 metadata query.
Abstract translation: 提供一种利用元数据搜索多媒体的设备和方法，可以方便地搜索多媒体数据，而无需使用MPEG-7查询格式的查询属性直接接收来自用户的MPEG（运动图像专家组）-7元数据中的查询 以及映射MPEG-7元数据的信息。 映射信息存储部分（40）存储/管理MPEG-7查询属性和MPEG-7元数据项之间的映射信息。 查询属性映射器（20）通过使用映射信息，根据用户查询获得映射到MPEG-y查询属性的MPEG-7元数据项。 查询输入部分（10）根据来自用户查询的MPEG-7查询格式生成并输出MPEG-7查询。 查询属性映射器通过使用映射的MPEG-7元数据项生成并输出MPEG-7元数据查询。 搜索器（30）通过使用MPEG-7元数据查询来搜索多媒体。
-
公开(公告)号：KR100641053B1
公开(公告)日：2006-11-02
申请号：KR1020050093880
申请日：2005-10-06
Applicant: 한국전자통신연구원
IPC: G06F17/27
Abstract: A device and a method for restoring an omitted component of a sentence are provided to prevent an error caused from an ellipsis of the sentence component, offer correct sentence structure analysis information, and recognize/restore the omitted component of the Hangul sentence by properly using rule and statistics information. A sentence structure(10) analyzer analyzes a structure of the inputted sentence based on predefined grammar. An ellipsis candidate recognizer(20) detects restoration candidates of the ellipsis in the inputted sentence if the ellipsis of the sentence is determined by checking a necessary component of each inflected word appeared in the analyzed sentence. An ellipsis restorer(30) restores the ellipsis in the detected restoration candidates by using the predefined rule/statistics information(32).
Abstract translation: 提供一种用于恢复句子的省略分量的设备和方法，以防止由句子分量的省略引起的错误，提供正确的句子结构分析信息，并且通过适当地使用规则来识别/恢复韩文句子的省略分量 和统计信息。 句子结构（10）分析器基于预定义语法分析输入句子的结构。 如果通过检查分析的句子中出现的每个屈曲词的必要成分来确定句子的省略号，则省略号候选识别器（20）检测输入语句中的省略号的恢复候选项。 省略号恢复器（30）通过使用预定义的规则/统计信息（32）来恢复检测到的恢复候选中的省略号。
-
公开(公告)号：KR1020060069616A
公开(公告)日：2006-06-21
申请号：KR1020040108121
申请日：2004-12-17
Applicant: 한국전자통신연구원
CPC classification number: G10L15/1822 , G10L2015/027 , Y10S707/99936
Abstract: An apparatus and a hybrid method for recognizing answer type are disclosed. The apparatus includes: a morpheme analyzer for analyzing morphemes of an input text; a syllabic answer type recognizer for extracting a predetermined size syllable from a morpheme list and recognizing an answer type based on the extracted syllable; a vocabulary feature recognizer for allocating feature to each morpheme and recognizing the feature; a vocabulary feature disambiguation unit for disambiguating vocabulary feature ambiguity of morphemes having more than one feature; a pattern rule answer type recognizer for recognizing an answer type by comparing a consecutive sequence of the morphemes and a consecutive sequence of constitutional features connected to the morphemes with a pre-constructed pattern rules; a statistic answer type recognizer for recognizing an answer type by implementing a statistic model; and an answer type sub-category recognizer for recognizing a sub-category of the recognized answer type classified to general category.
Abstract translation: 本发明涉及一种用于识别问答系统的韩语的正确答案类型的混合正确答案类型识别装置和方法。 本发明音节识别所提取的音节的基础上的答案类型提取形态学分析单元，一个预定尺寸的音节的形态列表由词素分析单元的每个分析用于分析输入文本的词素 基于该答案类型识别部和由每个词素列表的禀赋梗所分配的配置，关于具有一个或多个质量的词素由词法品质识别器识别出识别所述质量识别词汇的相应的词汇品质 和性别减轻词汇质量的一部分，以消除质量的属性，基于规则图案答案类型来识别正确的类型，以连接到具有从相对于既定的模式规则词素组词素一系列列表的结构素质的连续序列 一个识别单元，以及与连续列表和词素的语素连接的组成特征的连续列表 它是由用于通过将统计模型，其中，由所述识别单元识别的正确类型的详细标准答案类型识别的基于统计的答案类型正确类型的详细类别被识别为大类别识别正确的类型基于统计的答案类型识别部构成。
-
-
-
-
-
-
-
-
-
Search Results
123
records
(took 0.028 seconds)
