-
公开(公告)号：KR101309839B1
公开(公告)日：2013-09-23
申请号：KR1020090118298
申请日：2009-12-02
Applicant: 한국전자통신연구원
Abstract: 본 발명은 통계 정보를 이용한 규칙 기반 구문 분석 장치 및 방법에 관한 것으로, 본 발명의 일실시 예에 따른 통계정보를 이용한 규칙 기반 구문분석 방법은, 입력 문장에 대해 구문 규칙을 적용함으로써 구문 분석을 수행하는 단계; 상기 입력 문장에 대해 적용되는 규칙에 주어진 규칙 확률과 어휘통계정보에 기반하여 계산된 어휘 의존 가중치를 이용하여 상기 규칙에 대한 규칙 가중치를 계산하는 단계; 각 구문트리에 사용된 규칙에 대해 계산된 상기 규칙 가중치들을 곱하여 각 구문트리의 가중치를 계산하고 가장 높은 가중치를 갖는 구문 트리를 선택하는 단계; 및 상기 선택된 구문 트리를 출력하는 단계를 포함한다.
상술한 바와 같은 본 발명은, 규칙기반 방식의 효율성과 통계기반 방식의 높은 모호성 처리 성능을 갖는 구문분석이 가능하다.
언어 처리, 구문 분석, 통계 정보, 규칙 기반-
公开(公告)号：KR1020130042822A
公开(公告)日：2013-04-29
申请号：KR1020110106928
申请日：2011-10-19
Applicant: 한국전자통신연구원
IPC: G06F17/28
CPC classification number: G06F17/2881 , G06F17/2775
Abstract: PURPOSE: A translation device using a sample sentence based translation method and a pattern based translation method and a method thereof are provided to improve a translation performance by generating a translated sentence by comparing a sample sentence based translated sentence with a pattern based translated sentence. CONSTITUTION: An original sentence analysis unit(110) analyzes a sentence structure of an original sentence. A first translated sentence generation unit(120) translates phrases included in an original sentence based on a translation dictionary, a translation pattern, and a translation rule. A first translated sentence generation unit generates a first translated sentence corresponding to each phrase. A second translated sentence generation unit(130) translates the phrases included in the original sentence based on a phrase unit translation sample sentence. The second translated sentence generation unit generates a second translation sentence corresponding to each phrase. A translated sentence selection unit(150) selects the first translated sentence or the second translated sentence based on the reliability of the first and the second translated sentence. A translated sentence generation unit(160) generates a translated sentence based on the selected translated sentence. [Reference numerals] (100) Translation device; (110) Original sentence analysis unit; (120) Pattern-based translated sentence generation unit; (125) Translation information database; (130) Translation example-based translated sentence generation unit; (135) Translation example information database; (140) Translation example management unit; (150) Translated sentence selection unit; (160) Translated sentence generation unit;
Abstract translation: 目的：提供使用基于样本语言的翻译方法和基于模式的翻译方法及其方法的翻译装置，通过将基于翻译句子的翻译句子与基于模式的翻译句子进行比较来产生翻译句子来改善翻译性能。 构成：原句分析单元（110）分析原句的句子结构。 第一翻译句子生成单元（120）基于翻译字典，翻译模式和翻译规则翻译包含在原始句子中的短语。 第一翻译句子生成单元生成与每个短语相对应的第一翻译句子。 第二翻译句子生成单元（130）基于短语单元翻译样本语句来翻译原始句子中包含的短语。 第二翻译语句生成单元生成与每个短语对应的第二翻译句子。 翻译句子选择单元（150）基于第一和第二翻译句子的可靠性来选择第一翻译句子或第二翻译句子。 翻译句子生成单元（160）基于所选择的翻译句子生成翻译句子。 （附图标记）（100）翻译装置; （110）原句分析单位; （120）基于模式的翻译语句生成单元; （125）翻译信息数据库; （130）翻译实例翻译句子生成单元; （135）翻译示例信息数据库; （140）翻译示例管理单位; （150）翻译句子选择单位; （160）翻译句子生成单元;
-
公开(公告)号：KR1020120089502A
公开(公告)日：2012-08-13
申请号：KR1020100125870
申请日：2010-12-09
Applicant: 한국전자통신연구원
CPC classification number: G06F17/289 , G06F17/2809 , G06F17/27 , G06F17/271 , G06F17/2755 , G06F17/28 , G06F17/2854
Abstract: PURPOSE: A translation knowledge server generating method and an apparatus thereof are provided to obtain a translation knowledge and to apply the obtained knowledge to a translation engine. CONSTITUTION: A data analysis unit(103) performs morphological analysis and syntax analysis of initial translation knowledge data collected from a data collecting unit. The data analysis unit outputs the analyzed data. A translation knowledge learning unit(105) determines a target word by domain according to predetermined domain information. The translation knowledge learning unit determines a domain through automatic leaning clustering. The translation knowledge learning unit learns translation knowledge in real time.
Abstract translation: 目的：提供翻译知识服务器生成方法及其装置，以获得翻译知识并将获得的知识应用于翻译引擎。 构成：数据分析单元（103）执行从数据收集单元收集的初始翻译知识数据的形态分析和语法分析。 数据分析单元输出分析数据。 翻译知识学习单元（105）根据预定域信息确定目标单词。 翻译知识学习单元通过自动倾斜聚类来确定域。 翻译知识学习单元实时学习翻译知识。
-
公开(公告)号：KR1020120088032A
公开(公告)日：2012-08-08
申请号：KR1020100101419
申请日：2010-10-18
Applicant: 한국전자통신연구원
IPC: G06F17/28
CPC classification number: G06F17/28 , G06F17/2818 , G06F17/2854 , G06F17/2872 , G06F17/289
Abstract: PURPOSE: An automatic real time translation knowledge extracting/verifying method and an apparatus thereof are provided to automatically build a translation knowledge of an automatic translation system and to reduce installation costs for building existing automatic translation knowledge dictionary. CONSTITUTION: A parallel corpus collecting unit(100) collects a parallel corpus on a web. The parallel corpus collecting unit removes a tag such as HTML. The parallel corpus collecting unit generates the parallel corpus. A translation knowledge extracting block(110) extracts predetermined translation knowledge. A translation knowledge evaluating block(130) evaluates and converts the extracted translation knowledge into a first evaluation element and a second evaluation element.
Abstract translation: 目的：提供自动实时翻译知识提取/验证方法及其装置，以自动构建自动翻译系统的翻译知识，并降低构建现有自动翻译知识词典的安装成本。 构成：平行语料库收集单元（100）在网上收集平行语料库。 平行语料库收集单元删除诸如HTML的标签。 平行语料库收集单元生成平行语料库。 翻译知识提取块（110）提取预定的翻译知识。 翻译知识评估块（130）评估并将所提取的翻译知识转换为第一评估元素和第二评估元素。
-
公开(公告)号：KR1020110071205A
公开(公告)日：2011-06-29
申请号：KR1020090127709
申请日：2009-12-21
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2854
Abstract: PURPOSE: An automatic interpretation device and method is provided to enhance the interpreting performance by using/calculating the similarity between an interpreted sentence and a first language sentence which is a speech recognition result. CONSTITUTION: A speech recognizer(S100) receives the first language speech and generates the first language sentence with speech recognizing process. A language processor(S110) extracts the element included in the first language sentence. A similarity calculation part(S120) compares the extracted element and the element included in the interpreted sentence, and calculates the similarity between the first language sentence and the interpreted sentence. A sentence translation part(S130) translates the first language sentence into the second language sentence according to the similarity. A speech composing part(S140) extracts/composes the speech data corresponding to the second language sentence and outputs the speech signal.
Abstract translation: 目的：提供一种自动解释装置和方法，通过使用/计算解释语句和作为语音识别结果的第一语言句子之间的相似度来增强解释性能。 构成：语音识别器（S100）接收第一语言语音，并产生具有语音识别处理的第一语言句子。 语言处理器（S110）提取包含在第一语言句子中的元素。 相似度计算部分（S120）比较提取的元素和解释语句中包含的元素，并计算第一语言句和解释句之间的相似度。 句子翻译部分（S130）根据相似度将第一语言句子翻译成第二语言句子。 语音合成部（S140）提取/构成与第二语句对应的语音数据，并输出该语音信号。
-
Patent Agency Ranking
公开(公告)号：KR1020110067276A
公开(公告)日：2011-06-22
申请号：KR1020090123802
申请日：2009-12-14
Applicant: 한국전자통신연구원
CPC classification number: G06F17/278 , G06F17/2755 , G06F17/2863
Abstract: PURPOSE: A Korean language analyzing apparatus is provided to improve the efficiency of analyzing a noun phrase structure by determining noun phrase analysis candidates on the basis of a noun vocabulary pattern and a verb phrase pattern. CONSTITUTION: The inflected word expression/vocabulary pattern building block [verb phrase/vocabulary pattern building block](102) uses the sentences of the analysis object domain. The inflected word expression pattern and vocabulary pattern are away driven. A verb phrase/vocabulary pattern building block builds verb phrase patterns and vocabulary patterns using the sentences in a domain to be analyzed. An analysis candidate generating block(104) generates noun phrase analysis candidates by searching for a noun that can be a declinable word from the noun phrase sentence to be analyzed. A priority determining block(108) produces a preference from the established vocabulary patterns and determines the priority of the noun phrase analysis candidates. An analysis candidate selecting block(110) determines on noun phrase analysis candidate in consideration of the preference and inter-noun restriction.
Abstract translation: 目的：提供韩语分析装置，以通过基于名词词汇模式和动词短语模式确定名词短语分析候选来提高分析名词短语结构的效率。 构成：词汇表达/词汇模式构建块[动词短语/词汇模式构建块]（102）使用分析对象域的句子。 转折词表达模式和词汇模式被驱逐。 动词短语/词汇模式构建块使用要分析的域中的句子构建动词短语模式和词汇模式。 分析候选生成块（104）通过从要分析的名词短语句中搜索可以是可拒绝词的名词来生成名词短语分析候选。 优先级确定块（108）从所建立的词汇模式产生偏好，并确定名词短语分析候选者的优先级。 考虑到偏好和名词之间的限制，分析候选选择块（110）确定名词短语分析候选者。
-
公开(公告)号：KR1020110062867A
公开(公告)日：2011-06-10
申请号：KR1020090119720
申请日：2009-12-04
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30539 , G06F17/2735 , G06F17/2755 , G06F17/2863
Abstract: PURPOSE: A source language-object language term list constructing apparatus and method thereof are provided to construct an original language-object language term list through mining of a search result through an intermediate language. CONSTITUTION: A document search unit(105) collects a search result including an original language. A data mining unit(107) extracts intermediate language parallel sentence or a word. A morpheme analysis unit(109) analyzes an extracted original language-middle language parallel sentence. A word alignment unit(111) aligns the language parallel sentence. A term generator(113) generates an original language term list.
Abstract translation: 目的：提供源语言对象语言词列表构造装置及其方法，以通过中间语言挖掘搜索结果来构建原始语言对象语言词列表。 构成：文件搜索单元（105）收集包括原始语言的搜索结果。 数据挖掘单元（107）提取中间语言并行语句或单词。 语素分析单元（109）分析提取的原始语言 - 中间语言并行语句。 字对齐单元（111）对齐语言并行句。 术语生成器（113）生成原始语言项列表。
-
公开(公告)号：KR1020110057632A
公开(公告)日：2011-06-01
申请号：KR1020090114108
申请日：2009-11-24
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2863 , G06F17/218 , G06F17/271 , G06F17/2785
Abstract: PURPOSE: A Chinese translation apparatus and method thereof which uses syntax analysis information are provided to determine a Chinese sentence pattern by using syntax analysis information of a tagged Chinese sentence. CONSTITUTION: An input unit(102) performs segmentation and tagging of Chinese sentence and analyzes a syntax. A sentence pattern determining unit(104) uses a sentence pattern decision rule about the sentence analysis result. A transform unit(112) converts a Chinese sentence according to the sentence type. A terminative ending generating unit(114) generates terminative ending production rule corresponding to a sentence type.
Abstract translation: 目的：提供一种使用语法分析信息的中文翻译装置及其方法，通过使用标记中文句子的语法分析信息来确定中文句型。 构成：输入单元（102）执行中文句子的分割和标记，并分析语法。 句子模式确定单元（104）使用关于句子分析结果的句子模式决定规则。 变换单元（112）根据句型转换中文句子。 终端生成单元（114）生成与句型对应的终止生成规则。
-
公开(公告)号：KR1020110057495A
公开(公告)日：2011-06-01
申请号：KR1020090113923
申请日：2009-11-24
Applicant: 한국전자통신연구원
Abstract: PURPOSE: A method and device for segmenting a Chinese language are provided to phrase analysis accuracy and phrase analysis performance of Chinese based on phrase rule by performing a phrase segmentation based on Chinese contextual information. CONSTITUTION: A segmentation parameter setting unit(100) establishes a segment parameter about a sentence of Chinese in which morpheme is analyzed. A segment location estimating unit(102) presumes segment-available location and a maximum segment number about a Chinese sentence which the segment parameter is set.
Abstract translation: 目的：通过基于中文语境信息进行短语分割，提供了一种基于短语规则的汉语分词准确度和短语分析性能的方法和设备。 构成：细分参数设定单元（100）建立关于语素被分析的汉语句子的分段参数。 段位置估计单元（102）假设段可用位置和关于段参数被设置的中文句子的最大段号。
-
公开(公告)号：KR1020110028123A
公开(公告)日：2011-03-17
申请号：KR1020090086064
申请日：2009-09-11
Applicant: 한국전자통신연구원
CPC classification number: G06F17/289 , G06K9/2081 , G06K2209/01
Abstract: PURPOSE: An automatic translation apparatus and a method thereof using a user interaction of a mobile apparatus which modifies the error of a text are provided to increase the level of an automatic translation result and to modify the recognition error about the text through user feedback. CONSTITUTION: An text recognition controller(103) designates a target character string area through a user interaction in a digital image and performs the character recognition through an OCR(Optical Character Recognition). A text transmission controller(105) transmits the text character string. An automatic translation controller(107) generates a translation of the character string by using automatic translation knowledge database.
Abstract translation: 目的：提供一种使用修改文本错误的移动装置的用户交互的自动翻译装置及其方法，以增加自动翻译结果的级别，并通过用户反馈修改关于文本的识别错误。 构成：文本识别控制器（103）通过数字图像中的用户交互来指定目标字符串区域，并通过OCR（光学字符识别）执行字符识别。 文本传输控制器（105）发送文本字符串。 自动翻译控制器（107）通过使用自动翻译知识数据库生成字符串的翻译。
-
-
-
-
-
-
-
-
-
Search Results
199
records
(took 0.029 seconds)
