-
公开(公告)号：KR1020110061788A
公开(公告)日：2011-06-10
申请号：KR1020090118298
申请日：2009-12-02
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2715 , G06F17/271
Abstract: PURPOSE: A rule base syntax analyzing device and method thereof are provided to perform syntax analysis with a high processing performance and an efficiency of rule based method by processing ambiguity based on vocabulary dependence information from context tree attached corpus. CONSTITUTION: A rule bases parsing module(103) selects optimal context tree by performing syntax analyzing of input sentence based on syntax rule. A rule weight calculating module(105) calculates rule weight and provides the weight to a rule based parsing module by using vocabulary weight and rule probability of a rule applied about the input sentence.
Abstract translation: 目的：提供规则库语法分析设备及其方法，通过基于来自上下文树附加语料库的词汇依赖信息，通过处理模糊度，以高处理性能和基于规则的方法的效率来执行语法分析。 规则：规则库解析模块（103）通过基于语法规则执行输入句子的语法分析来选择最佳上下文树。 规则权重计算模块（105）通过使用关于输入句子的规则的词汇权重和规则概率来计算规则权重并将权重提供给基于规则的解析模块。
-
公开(公告)号：KR1020110061209A
公开(公告)日：2011-06-09
申请号：KR1020090117787
申请日：2009-12-01
Applicant: 한국전자통신연구원
CPC classification number: G06F17/289 , G06F17/2715 , G06F17/2755
Abstract: PURPOSE: A post-processing knowledge generation apparatus is provided to improve translation performance by correcting faults on translation based on post-processing knowledge. CONSTITUTION: An original text extracting unit(204) extracts an original text from a parallel corpus. A machine translation part(206) machine-translates the original text and creates machine translation corpus. An auto arranging part(210) arranges the machine translation corpus and a correct translation corpus which is extracted from the parallel corpus based on statistics. An extracting unit(212) extracts text arranging information by the arranging result. A filter(214) amends the error of the text arranging information and creates post-processing knowledge.
Abstract translation: 目的：提供后处理知识产生装置，通过基于后处理知识修正翻译故障，提高翻译性能。 构成：原始文本提取单元（204）从平行语料库中提取原始文本。 机器翻译部分（206）机器翻译原始文本并创建机器翻译语料库。 自动排列部件（210）根据统计信息，从平行语料库中排列机器翻译语料库和正确的翻译语料库。 提取单元（212）通过排列结果提取文本排列信息。 过滤器（214）修正文本排列信息的错误，并创建后处理知识。
-
193.发明授权
公开(公告)号：KR101027007B1
公开(公告)日：2011-04-11
申请号：KR1020080131758
申请日：2008-12-22
Applicant: 한국전자통신연구원
IPC: G06F17/40
Abstract: 본 발명은 자동 번역 시스템의 도메인 변화에 따른 대역어 사전의 특화 기법에 관한 것으로, 목표 도메인에 속하는 원시 언어 코퍼스와 목표 언어 코퍼스를 이용하여 공기 어휘를 추출하고, 이를 대역어 사전에 매핑시켜 대역어 후보를 추출하며, 이에 대한 대역 관계의 오류를 필터링한 후 대표 대역어를 결정하여 대역어 사전에 반영함으로써, 자동 번역 시스템의 대역어 사전을 자동으로 특화시킬 수 있어 이를 구축하는데 소요되는 비용을 절감할 수 있는 것이다.
자동 번역, 대역어 사전-
公开(公告)号：KR101027518B1
公开(公告)日：2011-04-06
申请号：KR1020080131764
申请日：2008-12-22
Applicant: 한국전자통신연구원
IPC: G10L15/193 , G06F17/27 , G06F17/28
Abstract: 본 발명은 가중 유한 상태 변환기(Weighted Finite State Transducer)에서, 입력 기호가 널(null)인 천이, 예컨대 ε(epsilon ; 엡실론) 입력 천이를 제거하는 방법에 관한 것이다. 가중 유한 상태 변환기의 등가성(equivalency)을 유지하면서, 각 상태(state, 또는 노드(node))의 종류에 따라 ε 입력 천이 제거 방법을 달리하여 가중 유한 상태 변환기의 크기를 최소화하는 방향으로 ε 입력 천이를 제거한다.
가중 유한 상태 변환기, 널 심볼 천이-
公开(公告)号：KR101025814B1
公开(公告)日：2011-04-04
申请号：KR1020080127710
申请日：2008-12-16
Applicant: 한국전자통신연구원
Abstract: 본 발명은 운율 모델을 이용한 형태소 품사 태깅 방법 및 그 장치에 관한 것으로, 음성인식의 결과 또는 전사한 텍스트문장의 형태소를 해석하고, 해석이 모호한 부분에 대해 음성 DB를 기반으로 기 구축된 형태소 품사 시퀀스별 운율 모델과 입력음성의 운율 모델을 비교하여, 입력음성의 운율 모델에 대한 최적의 형태소 품사 시퀀스를 찾고, 그 찾은 결과를 형태소 품사 태깅 방법과 조합시켜 형태소 품사를 태깅함으로써, 형태소 품사 태깅의 정확도를 극대화시킬 수 있다. 또한, 본 발명은 운율 모델을 적용하여 형태소 품사를 태깅함으로써, 화자의 발성 의도를 파악할 수 있다.
형태소, 운율 모델, 품사, 태깅-
Patent Agency Ranking
公开(公告)号：KR1020110018140A
公开(公告)日：2011-02-23
申请号：KR1020090075778
申请日：2009-08-17
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2705 , G06F17/18 , G06F17/2755 , G06F17/277
Abstract: PURPOSE: A statistical HMM part tagging apparatus and a method thereof for increasing the tagging performance of a morpheme part are provided to increase the accuracy of tagging about a document of a domain by using learning information. CONSTITUTION: A real time learning based statistical part tagging unit(103) utilizes vocabulary stored in a vocabulary probability information DB and context stored in a context probability information DB. A real time learning based tagging error correction unit(105) corrects an error through a tagging error modification DB. A real time document information learning unit(101) establishes real time context probability information DB after extracting context probability information.
Abstract translation: 目的：提供用于提高语素部分的标记性能的统计HMM部分标签装置及其方法，以通过使用学习信息来增加关于域的文档的标签的准确性。 构成：基于实时学习的统计部分标记单元（103）利用存储在词汇概率信息DB中的词汇和存储在上下文概率信息DB中的上下文。 基于实时学习的标记纠错单元（105）通过标记错误修改DB来校正错误。 实时文档信息学习单元（101）在提取上下文概率信息之后建立实时上下文概率信息DB。
-
公开(公告)号：KR1020110017600A
公开(公告)日：2011-02-22
申请号：KR1020090075145
申请日：2009-08-14
Applicant: 한국전자통신연구원
Abstract: PURPOSE: An apparatus for searching a word entry in a portable electronic dictionary and a method thereof are provided to output N-best recognition results and enables a user to select one of the results when performing a dictionary searching operation for a foreign language through a voice recognition technology. CONSTITUTION: A pre-treatment unit receives a voice signal for the combination of continuous pronunciation of each letter which configures a dictionary pronunciation or a word, and a word network configuration unit(512) receives a pronunciation string from the stored multi-pronunciation dictionary information to configure a network through the matching with the extracted phoneme series. A searching unit(514) refers to a triphone unit acoustic model transferred through a training unit and the configured network in order to search the word corresponding to the voice signal.
Abstract translation: 目的：提供一种用于搜索便携式电子词典中的词条的装置及其方法，用于输出N最佳识别结果，并且使用户能够通过语音执行外语的字典搜索操作时选择结果之一 识别技术。 构成：预处理单元接收用于组合字典发音或单词的每个字母的连续发音的组合的语音信号，并且字网络配置单元（512）从存储的多发音字典信息中接收发音串 通过与提取的音素系列匹配来配置网络。 搜索单元（514）是指通过训练单元和配置的网络传送的三音单元声学模型，以搜索对应于语音信号的单词。
-
公开(公告)号：KR1020100073182A
公开(公告)日：2010-07-01
申请号：KR1020080131776
申请日：2008-12-22
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2705 , G06F17/2755
Abstract: PURPOSE: A neologism selection device and a method thereof are provided to determine priority of neologism candidate based on a theme and select neologism according to the priority. CONSTITUTION: A morpheme analyzer(102) performs morphological analysis about input web document. A keyword extractor(106) extracts a keyword corresponding to neologism candidate from analyzed sentence. A subject detecting and tracking unit(108) performs theme detection and theme tracking through the extracted keyword. According to a subject keyword weight of the keyword, a priority determination unit(110) selects neologism according to the priority after determination of priority.
Abstract translation: 目的：提供新词选择装置及其方法，以根据主题确定新词候选者的优先级，并根据优先级选择新词。 构成：语素分析器（102）对输入的Web文档进行形态分析。 关键词提取器（106）从分析语句中提取与新词候选者相对应的关键字。 被摄体检测跟踪单元（108）通过提取的关键字进行主题检测和主题跟踪。 根据关键词的主题关键字权重，优先级确定单元（110）在确定优先级之后根据优先级选择新词。
-
公开(公告)号：KR1020100073170A
公开(公告)日：2010-07-01
申请号：KR1020080131764
申请日：2008-12-22
Applicant: 한국전자통신연구원
IPC: G10L15/193 , G06F17/27 , G06F17/28
Abstract: PURPOSE: A null symbol transition removal method in weighting finite state transducer is provided to remove all ε input transitions except a loop formed by ε input transitions, thereby minimizing the size of the weighting finite state transducer according to a form of input transitions of the whole weighted finite state transducer. CONSTITUTION: An input transition search unit(100) searches null symbol transition while searching the whole weighted finite state transducer, for example searching ε input transition. An input transition determination unit(102) determines type of the ε input transition. An input transition removing unit(104) removes the ε input transition according to type of the determined ε input transition. If an input symbol is null, forward direction processing suitable for node type is performed. If the input symbol is not null, the next transition is performed.
Abstract translation: 目的：提供加权有限状态传感器中的零符号跃迁去除方法，以除去除了ε输入跃迁形成的环之外的所有ε输入跃迁，从而根据整体输入转换的形式最小化加权有限状态换能器的大小 加权有限状态传感器。 构成：输入转换搜索单元（100）在搜索整个加权有限状态换能器时搜索零符号转换，例如搜索ε输入转换。 输入转换确定单元（102）确定ε输入转换的类型。 输入转换去除单元（104）根据确定的ε输入转换的类型去除ε输入转变。 如果输入符号为空，则执行适合于节点类型的正向处理。 如果输入符号不为空，则执行下一个转换。
-
公开(公告)号：KR1020100073162A
公开(公告)日：2010-07-01
申请号：KR1020080131756
申请日：2008-12-22
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2809 , G06F17/274 , G06F17/2755
Abstract: PURPOSE: A method and a device for generating translation sentence in an automatic English-Korean translation system are provided to automatically generate prefinal-ending corresponding to information delivered from Korean to English verb by information which an auxiliary verb or a temporal adverbial phrase transfers to a verb. CONSTITUTION: An English morpheme analyzer(100) analyzes input English sentence in a morpheme unit. A structure analyzer(300) analyzes English structure. English transforming unit(400) changes construction analysis result into Korean structure. A Korean generator(500) generates the Korean structure in the Korean morpheme unit. A perfinal-ending processor(600) transfers Korean perfinal-ending information corresponding to information about the English auxiliary verb group to the Korean generator.
Abstract translation: 目的：提供一种用于在自动英语 - 韩语翻译系统中生成翻译句子的方法和装置，用于根据辅助动词或时间语言短语转移到英文动词的信息自动生成对应于从韩语输入到英语动词的信息的前缀结尾 动词。 构成：英语词素分析器（100）分析语素单位中的输入英语句子。 结构分析器（300）分析英文结构。 英文变换单位（400）将建筑分析结果变更为韩国结构。 韩国发电机（500）在韩国语素单位中生成韩国结构。 一个付费终端处理器（600）将与英语辅助动词组相关的韩语完成信息传送给韩语发音器。
-
-
-
-
-
-
-
-
-
Search Results
434
records
(took 0.03 seconds)
