公开(公告)号：KR100276687B1

公开(公告)日：2001-01-15

申请号：KR1019980048384

申请日：1998-11-12

Applicant: 한국전자통신연구원

Inventor： 차건회 ,  박재득

IPC: G09B7/00

Abstract: 본 발명은 대화식 학습 보조 장치 및 그의 대화분석 방법에 관한 것이다.
본 발명은 형태소 및 구문 분석과, 의미 및 담화분석에 필요한 데이터를 제공하기 위한 사전저장 장치와, 대화분석에 필요한 데이터를 제공하기 위한 지식베이스 저장 장치와, 마이크를 통해 입력된 음성을 동일한 내용의 텍스트로 변환하기 위한 음성/텍스트 변환 장치와, 상기 음성/텍스트 변환 장치로부터 출력되는 데이터를 상기 사전저장 장치의 데이터와 비교하여 순차적으로 형태소 분석, 구문 분석 및 의미 분석을 하기 위한 형태소 해석기, 구문 분석기 및 의미 해석기와, 상기 의미 해석기로부터 출력되는 데이터를 상기 사전저장 장치의 데이터를 참조하여 생략 및 대용어 해결을 하기 위한 담화 분석기와, 상기 담화 분석기로부터 출력되는 데이터를 상기 지식베이스 저장 장치의 데이터와 비교하여 용언의 하위범주화 정보로 변환하고, 이것� �로부터 화행, 영역내 키워드, 영역내 적합성 판단, 데이터베이스 질의어 생성 및 검색 등을 수행하기 위한 대화 관리기와, 상기 대화 관리기로부터 출력되는 데이터를 상기 음성/텍스트 변환 장치의 데이터와 비교하여 화행, 데이터베이스 검색 결과 및 영역내 적합도로부터 사용자에게 제공할 문장을 생성하기 위한 응답 생성기를 포함하여 이루어지며, 인간이 사용하는 언어만을 통하여 자신의 학습에 도움을 얻는 장치를 획득할 수 있는 장치 및 대화분석 방법을 제공한다.

公开(公告)号：KR1020000032056A

公开(公告)日：2000-06-05

申请号：KR1019980048384

申请日：1998-11-12

Applicant: 한국전자통신연구원

Inventor： 차건회 ,  박재득

IPC: G09B7/00

CPC classification number: G06F17/3043 ,  G06F17/271 ,  G06F17/2755 ,  G06F17/2785 ,  G06F17/30554 ,  G06N5/04 ,  G10L15/18

Abstract: PURPOSE: A conversatio -type learning auxiliary apparatus is provided to learn through a language used by a human without a technique or detailed knowledge of a computer. CONSTITUTION: A conversation-type learning auxiliary apparatus comprises a voice/text converting part(2) which converts a voice inputted through a microphone(1) into a text of the same contents. A morpheme parser(3) compares data from the converting part(2) with data of a dictionary storing device(10) to parse a morpheme, and a sentence structure parser(4) compares data from the parser(3) with data of a dictionary storing device(10) to parse a sentence structure. A meaning parser(5) compares data from the parser(4) with data of a dictionary storing device(10) to parse the meaning. Data from the parser(5) is transferred to a conversation manager(7) through a conversation parser(6). The conversation manager(7) compares the data from the conversation parser(6) with data of a knowledge base storing device(11) to convert a comparison result into lower category information of a declinable word. A response generator(8) compares data from the conversation manager(7) with the data from the voice/text converting part(2) to generate a sentence to be outputted through an output device(9).

Abstract translation: 目的：提供一种对话式学习辅助设备，以便在没有计算机的技术或细节知识的情况下学习人类使用的语言。 构成：对话式学习辅助装置包括将通过麦克风（1）输入的语音转换为相同内容的文本的语音/文本转换部分（2）。 语素解析器（3）将来自转换部分（2）的数据与词典存储装置（10）的数据进行比较以解析语素，并且句子结构解析器（4）将来自解析器（3）的数据与 字典存储装置（10），用于解析句子结构。 意义分析器（5）将来自解析器（4）的数据与字典存储设备（10）的数据进行比较以解析意义。 来自解析器（5）的数据通过会话解析器（6）传送到会话管理器（7）。 会话管理器（7）将来自会话解析器（6）的数据与知识库存储设备（11）的数据进行比较，以将比较结果转换为可下降字词的较低类别信息。 响应发生器（8）将来自会话管理器（7）的数据与来自语音/文本转换部分（2）的数据进行比较，以生成要通过输出设备（9）输出的句子。

公开(公告)号：KR1020020045343A

公开(公告)日：2002-06-19

申请号：KR1020000074768

申请日：2000-12-08

Applicant: 한국전자통신연구원

Inventor： 차건회 ,  정의석 ,  임수종 ,  강현규

IPC: G06F17/30

CPC classification number: G06F17/30684 ,  Y10S707/99932 ,  Y10S707/99933 ,  Y10S707/99935

Abstract: PURPOSE: A device and a method for generating/searching the information based on the standardized sentence and meaning structure are provided to enable a user to describe the information provided by the user and search the information desired by the user. CONSTITUTION: The device for generating/searching the information comprises a data storage(11) storing the language knowledge data, the meaning expression data and a web document, an information providing interface(12) receiving the sentence generation for a transaction and a natural language query from the user, an input sentence analyzer(13) fro generating the meaning structure by analyzing the natural language sentence inputted from the user, a dialog processor(14) outputting a result, an index and a search result by resolving the meaning when a spelling and spacing error and the ambiguities of the sentence and meaning structure is occurred, a meaning structure processor(15) indexing/storing/searching the meaning structure receiving from the input sentence analyzer, and an information demand interface(16) for providing the data output from the dialog processor to the user.

Abstract translation: 目的：提供一种基于标准化句子和含义结构生成/搜索信息的设备和方法，以使用户能够描述用户提供的信息并搜索用户期望的信息。 构成：用于生成/搜索信息的装置包括存储语言知识数据，意义表达数据和web文档的数据存储器（11），接收交易的语句生成和自然语言的信息提供接口（12） 通过分析从用户输入的自然语言语句生成意义结构的输入句子分析器（13），输出结果的对话处理器（14），通过解析结果的索引和搜索结果来解析 拼写和间隔误差以及句子和意义结构的模糊性，意义结构处理器（15）索引/存储/搜索从输入句子分析器接收的意义结构，以及用于提供数据的信息需求接口（16） 从对话处理器输出到用户。

公开(公告)号：KR100413966B1

公开(公告)日：2004-01-07

申请号：KR1020000077945

申请日：2000-12-18

Applicant: 한국전자통신연구원

Inventor： 정의석 ,  차건회 ,  임수종 ,  윤보현 ,  왕지현 ,  강현규

IPC: G06F17/27

Abstract: PURPOSE: A system and a method for guiding a standard sentence pattern are provided to guide a user in a standard sentence pattern in real time by consulting the standard Korean sentence pattern. CONSTITUTION: The system includes an input unit(110) through which a user inputs a sentence, a morpheme analyzing unit(120) to divide the inputted sentence into the syllables and to analyze the morphemes composing the syllables, a vocabulary information extracting unit(130) to extract the vocabulary information of the analyzed sentence, a standard sentence pattern rule determining unit(140) to determine standard sentence pattern rules and apply the rules, and an output unit(170) to output a standardized sentence pattern of the inputted sentence depending on the applied standard sentence pattern rules. The standard sentence pattern rule determining unit applies the most proper one among the Korean standard sentence pattern rules loaded and pre-stored in the system. A standard sentence pattern is verified and deducted by using the abbreviation deduction patterns sought by an abbreviation deduction pattern search unit(150).

Abstract translation: 目的：提供一种指导标准句型的系统和方法，通过查阅标准的韩语句型，实时引导用户使用标准句型。 该系统包括用户通过其输入句子的输入单元（110），将输入的句子划分成音节并分析组成音节的语素的词素分析单元（120），词汇信息提取单元（130 ）提取所分析的句子的词汇信息;标准句型规则确定单元（140），用于确定标准句型规则并应用规则;以及输出单元（170），用于输出所输入句子的标准化句型， 对应用的标准句型规则。 标准句型规则确定单元应用加载并预先存储在系统中的韩语标准句型规则中最适当的一个。 通过使用由缩写推断模式搜索单元（150）搜索的缩写推断模式来验证并且扣除标准句型。

公开(公告)号：KR1020020058639A

公开(公告)日：2002-07-12

申请号：KR1020000086754

申请日：2000-12-30

Applicant: 한국전자통신연구원

Inventor： 윤보현 ,  정의석 ,  차건회 ,  강현규 ,  왕지현

IPC: G06F17/30

CPC classification number: G06F17/30616 ,  G06F17/30663 ,  G06F17/3069 ,  G06F17/30911

Abstract: PURPOSE: A system for searching an XML document and a method thereof are provided to search contents and a structure with respect to a user's query integrally from indexed information by integrally indexing contents and a structure with respect to an XML document. CONSTITUTION: A DTD(Document Type Definition) reduction unit(200) is provided for reducing a complicated DTD to a simple DTD for being used in an index and a search and making an index config file. An index unit(210) is provided for receiving the config file and an XML document made at the DTD reduction unit(200) for an index. An index information storing unit(230) is provided for receiving and storing index information from the index unit(210). A search unit is provided for receiving and searching a general query and a structure query from a user. An index document converting unit(211) receives the XML document and the config file, performs a parsing of the XML document, and makes a file for an index. A morpheme interpreting unit(212) is provided for interpreting a morpheme with respect to an index file made by the index document converting unit(211). An index language extracting unit(213) for extracting an index language in a result of the morpheme interpreting unit(212). An element and position information extracting unit(214) is provided for extracting element information and position information of the index language extracted in the index language extracting unit(213).

Abstract translation: 目的：提供一种用于搜索XML文档的系统及其方法，用于通过对内容和关于XML文档的结构进行整体索引来从索引信息中整体地搜索关于用户查询的内容和结构。 规定：提供DTD（文件类型定义）缩小单元（200），用于将复杂的DTD减少到简单的DTD，用于索引和搜索以及制作索引配置文件。 提供索引单元（210），用于接收用于索引的配置文件和在DTD缩减单元（200）处制作的XML文档。 索引信息存储单元（230）用于从索引单元（210）接收和存储索引信息。 提供搜索单元用于从用户接收和搜索一般查询和结构查询。 索引文档转换单元（211）接收XML文档和配置文件，执行XML文档的解析，并创建索引文件。 语素解释单元（212）被提供用于相对于由索引文档转换单元（211）制作的索引文件来解释语素。 一种用于提取语素解释单元（212）的结果中的索引语言的索引语言提取单元（213）。 提供元素和位置信息提取单元（214），用于提取在索引语言提取单元（213）中提取的索引语言的元素信息和位置信息。

公开(公告)号：KR1020020048715A

公开(公告)日：2002-06-24

申请号：KR1020000077945

申请日：2000-12-18

Applicant: 한국전자통신연구원

Inventor： 정의석 ,  차건회 ,  임수종 ,  윤보현 ,  왕지현 ,  강현규

IPC: G06F17/27

Abstract: PURPOSE: A system and a method for guiding a standard sentence pattern are provided to guide a user in a standard sentence pattern in real time by consulting the standard Korean sentence pattern. CONSTITUTION: The system includes an input unit(110) through which a user inputs a sentence, a morpheme analyzing unit(120) to divide the inputted sentence into the syllables and to analyze the morphemes composing the syllables, a vocabulary information extracting unit(130) to extract the vocabulary information of the analyzed sentence, a standard sentence pattern rule determining unit(140) to determine standard sentence pattern rules and apply the rules, and an output unit(170) to output a standardized sentence pattern of the inputted sentence depending on the applied standard sentence pattern rules. The standard sentence pattern rule determining unit applies the most proper one among the Korean standard sentence pattern rules loaded and pre-stored in the system. A standard sentence pattern is verified and deducted by using the abbreviation deduction patterns sought by an abbreviation deduction pattern search unit(150).

Abstract translation: 目的：提供一种用于指导标准句型的系统和方法，通过咨询标准的韩语句子模式实时指导用户的标准句型。 构成：系统包括用户输入句子的输入单元（110），语素分析单元（120），将输入的句子划分成音节并分析构成音节的语素，词汇信息提取单元（130） ）提取分析句子的词汇信息，用于确定标准句型规则并应用规则的标准句型图案规则确定单元（140），以及输出单元（170），以输出所输入的句子的标准化句型依赖 对应用标准句型规则。 标准句型规则确定单元在系统中加载和预存的韩文标准句模式规则中应用最合适的一种。 通过使用由缩写扣除模式搜索单元（150）寻求的缩写扣除模式来验证和扣除标准句型。

公开(公告)号：KR1020010008962A

公开(公告)日：2001-02-05

申请号：KR1019990027068

申请日：1999-07-06

Applicant: 한국전자통신연구원

Inventor： 차건회 ,  박재득 ,  이현아 ,  채영숙

IPC: G06F17/30

Abstract: PURPOSE: An apparatus and method for searching information using a concept classification network is provided to search desired information for a user with a natural language sentence. CONSTITUTION: A multimedia information natural language search apparatus is comprised of a natural language sentence input device(101), an input sentence analysis part(11), a data storage part(12), a search processing part(13) and an output device(108). The input sentence analysis part(11) consists of a morpheme analyzer(102), a sentence structure analyzer(103) and a semantic analyzer(104). The data storage part(12) consists of a dictionary storage device(109), a concept classification network storage device(110) and a multimedia database(111). The search processing part(13) consists of a semantic information extractor(105), a record searcher(106) and a response generator(107). If the natural language sentence input device(101) transmits a search condition to express through a natural language sentence inputted from a user to the input sentence analysis part(11), the input sentence analysis part(11) compares a sentence inputted from the user with the data of the dictionary storage device(109) and analyzes it through the morpheme analyzer(102), the sentence structure analyzer(103) and the semantic analyzer(104) sequentially. The semantic information extractor(105) of the search processing part(13) compares the data output from the semantic analyzer(104) with the data of the concept classification network storage device(110) and extracts a semantic information value.

Abstract translation: 目的：提供一种使用概念分类网络搜索信息的装置和方法，用于搜索具有自然语言句子的用户的期望信息。 构成：多媒体信息自然语言搜索装置由自然语言句子输入装置（101），输入句子分析部分（11），数据存储部分（12），搜索处理部分（13）和输出装置 （108）。 输入句子分析部分（11）由语素分析器（102），句子结构分析器（103）和语义分析器（104）组成。 数据存储部分（12）由词典存储装置（109），概念分类网络存储装置（110）和多媒体数据库（111）组成。 搜索处理部分（13）由语义信息提取器（105），记录搜索器（106）和响应发生器（107）组成。 如果自然语言句子输入装置（101）将通过从用户输入的自然语言句子的搜索条件发送到输入句子分析部分（11），则输入句子分析部分（11）比较从用户输入的句子 与字典存储装置（109）的数据一起通过语素分析器（102），句子结构分析器（103）和语义分析器（104）进行分析。 搜索处理部分（13）的语义信息提取器（105）将从语义分析器（104）输出的数据与概念分类网络存储设备（110）的数据进行比较，并提取语义信息值。