Abstract:
PURPOSE: A conversation method and system therefor are provided to enable a user to control system articulation step according to user learning steps by controlling various articulation flows. CONSTITUTION: A voice recognition unit(102) changes the received articulation of a user into articulation text by using articulation information. A language understanding unit(103) determines the articulation action of the user by using the changed articulation text. A conversation and progress management unit(104) determines one articulation point from the articulation points connected a target articulation point. A system conversation creation unit(106) searches articulation patterns connected with the articulation point corresponding to the determined articulation point.
Abstract:
PURPOSE: A method and apparatus for analyzing syntax is provided to accurately extract lexical information by the use of target words. CONSTITUTION: Apparatus for analyzing syntax comprises: a receiving part receiving a first sentence structure analysis result and a second sentence structure analysis result; a lexical information extraction unit(101) extracting translation word of translation target language corresponding with LDLex(Long Distance Lexeme) by the use of a first sentence structure analysis result, extracting a dependent word corresponding with the translation word by the use of the second sentence analysis result, and a step of extracting the dependant word from translation target input sentence by the use of the first sentence analysis result.
Abstract:
PURPOSE: An automatic translation pattern data construction method and an apparatus thereof are provided to apply obtained translation pattern data regardless of an automatic translation engine and to promote the performance of translation by correcting a weak point of the automatic translation system. CONSTITUTION: A pair of inputted double language corpus sentences is divided into a source language and a target language(120). The source language is generated as a test target language. The source language is collected based on the similarity of the test target language and the target language.
Abstract:
본 발명은 자동번역 시스템의 번역 오류를 후처리로 보정하여 번역의 품질을 향상시키는 번역 오류 후처리 보정 기술에 관한 것으로, 목적 언어 코퍼스로부터 번역 오류 유형에 특화된 언어 모델을 구축하고, 오류 특화 언어 모델에 기반하여 번역 오류를 탐색한 후, 오류보정 우선순위 결정 규칙에 따라 탐색된 번역 오류들 간의 오류 보정 우선순위를 정하고, 우선순위에 따라 차례로 탐색된 오류에 대한 보정 후보를 생성하고, 오류 특화 언어모델에 기반한 보정어 선택을 수행한 후, 번역 결과를 수정하는 과정을 반복함으로써 번역문에서 탐색된 모든 오류를 보정하는 것을 특징으로 한다. 본 발명에 의하면, 비문이나 자연스럽지 못한 표현 등과 같은 자동 번역 시스템의 번역 오류를 실시간으로 보정함으로써 자동 번역 시스템의 번역 성능을 향상시킬 수 있다. 자동 번역, 오류 유형 특화 언어 모델, 번역오류 보정
Abstract:
PURPOSE: A hybrid interpreting device and method thereof are provided to prevent ambiguity of an automatic interpreting by interpreting a primitive language sentence through hybrid scheme including the statistic based method and the pattern based method. CONSTITUTION: A source language input unit(102) generalizes an inputted primitive language sentence into a node unit. A first interpreting result generating unit(106) generates a primitive language which is generalized into the node unit as a first interpreted result which is converted into a node expression by using statistic based interpreting knowledge.
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:
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:
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