公开(公告)号：WO2010050638A1

公开(公告)日：2010-05-06

申请号：PCT/KR2008/007009

申请日：2008-11-27

Applicant: KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY ,  CHOI, Key-Sun ,  AHN, Jin hyun ,  MOON, Il Chul

Inventor： CHOI, Key-Sun ,  AHN, Jin hyun ,  MOON, Il Chul

IPC: G06F3/06 ,  G06F3/00 ,  G06F15/16

CPC classification number: G06F17/27 ,  G06F3/04883

Abstract: A machine-readable electronic memo system and method are provided. The machine-readable electronic memo system includes an electronic scratch paper, on which symbols or characters are written, for transmitting contents to a communication network and an electronic unit, an electronic memo input unit for inputting the symbols or characters on the electronic scratch paper, an electronic memo service unit including an electronic memo interpreting/processing part for creating a Triplet of RDF form from the contents of the electron scratch paper, an electronic memo inferring part for generating new information to be inferred from the Triplet based on a machine-readable knowledge structure, and a machine-readable knowledge structure, and an electronic memo network browser.

Abstract translation: 提供了一种机读电子备忘录系统和方法。 机读电子备忘录系统包括写入符号或字符的电子刮纸，用于将内容发送到通信网络和电子单元，用于在电子刮纸上输入符号或字符的电子备忘录输入单元， 电子备忘录服务单元，其包括电子备忘录解释/处理部分，用于从电子刮写纸的内容创建RDF格式的三重符号，电子备忘录推断部分，用于基于机器可读的从Triplet生成要被推断的新信息 知识结构和机器可读知识结构，以及电子备忘录网络浏览器。

公开(公告)号：WO2010058913A2

公开(公告)日：2010-05-27

申请号：PCT/KR2009006256

申请日：2009-10-28

Applicant: KOREA ADVANCED INST SCI & TECH ,  CHOI KEY-SUN ,  RYU PUM MO

Inventor： CHOI KEY-SUN

IPC: G06F17/00 ,  G06F17/30

CPC classification number: G06F17/30286

Abstract: The present invention relates to a method for automatically reasoning an "is-a" relationship, and a computer-readable recording medium containing a program for executing the method for automatically reasoning an "is-a" relationship in a computer. A method for reasoning an "is-a" relationship inherent in an "is-a" relationship list, comprises the steps of: converting the "is-a" relationship list into a description logic-based logical expression, and reasoning a new "is-a" relationship by using a reasoning engine.

Abstract translation: 本发明涉及一种用于自动推理“is-a”关系的方法，以及包含用于执行在计算机中自动推理“is-a”关系的方法的程序的计算机可读记录介质。 一种用于推理“is-a”关系列表中固有的“is-a”关系的方法包括以下步骤：将“is-a”关系列表转换为基于描述逻辑的逻辑表达，并推导新的“ 是使用推理引擎的“关系”。

公开(公告)号：WO2010050675A2

公开(公告)日：2010-05-06

申请号：PCT/KR2009005265

申请日：2009-09-16

Applicant: KOREA ADVANCED INST SCI & TECH ,  CHOI KEY-SUN ,  CHOI DONGHYUN

Inventor： CHOI KEY-SUN ,  CHOI DONGHYUN

IPC: G06F17/26

CPC classification number: G06F17/279

Abstract: The present invention relates to a method and to a system for automatically extracting relation triplet information for building ontology from texts with grammatically correct plain declarative sentences. The method for automatically extracting relation triplets through a dependency grammar parse tree according to the present invention comprises: a first step of performing syntactic parsing on an objective sentence to generate a dependency grammar parse tree; a second step of preprocessing the dependency grammar parse tree generated in the first step to convert the tree into a format suitable for relation extraction; and a third step of traversing, in a post-order manner, the dependency grammar parse tree converted in the second step to automatically extract relation triplets.

Abstract translation: 本发明涉及一种用于从具有语法正确的简单陈述句子的文本中自动提取建立本体的关系三元组信息的方法和系统。 根据本发明的通过依赖语法解析树自动提取关系三元组的方法包括：对客观句子执行句法解析以生成依赖性语法解析树的第一步骤; 在第一步骤中生成的依赖性语法分析树预处理的第二步骤，以将树转换成适合于关系提取的格式; 以及以顺序方式遍历在第二步骤中转换的依赖性语法解析树以自动提取关系三元组的第三步骤。

公开(公告)号：WO2009131417A3

公开(公告)日：2010-01-21

申请号：PCT/KR2009002165

申请日：2009-04-24

Applicant: KOREA ADVANCED INST SCI & TECH ,  CHOI KEY-SUN ,  CHOI DONGHYUN

Inventor： CHOI KEY-SUN ,  CHOI DONGHYUN

IPC: G06F17/27

CPC classification number: G06F17/2785

Abstract: The present invention relates to a sentence summarization method using a dependency grammar syntax tree, which creates a new dependency grammar syntax tree by modifying a dependency grammar syntax tree of a sentence according to a given rule, and then summarizes the sentence using the new syntax tree. A sentence summarization method using a dependence grammar syntax tree according to the invention comprises: a first step, in which a first dependency grammar syntax tree is created by using a syntax analyzer based on rules; a second step, in which a second dependency grammar syntax tree is created by using a syntax analyzer based on the probability for words without having a dependency relation therefor because of ambiguity in the first step; a third step, in which a third dependency grammar syntax tree is created by combining the first dependency grammar syntax tree created in the first step with the second dependency grammar syntax tree created in the second step; a fourth step, in which a modified fourth dependency grammar syntax tree is created by using and applying an artificial intelligence model that identifies a 'REMAIN' operation for maintaining a node of interest for each node from the third dependency grammar syntax tree, a 'DELETE' operation for deleting a node of interest and all of its child nodes, or a 'CHILD_UP' operation for deleting a node of interest and then putting its child node in its place; and a fifth step, in which a summary is created from the modified fourth dependency grammar syntax tree.

Abstract translation: 本发明涉及一种使用依赖语法语法树的句子汇总方法，该方法通过根据给定的规则修改句子的依赖语法语法树来创建新的依赖语法语法树，然后使用新的语法树来总结句子 。 使用根据本发明的依赖性语法语法树的句子汇总方法包括：第一步骤，其中通过使用基于规则的语法分析器来创建第一依赖性语法语法树; 第二步，其中通过使用语法分析器来创建第二依赖语法语法树，所述语法分析器基于由于第一步骤中的歧义而不具有依赖关系的单词的概率; 第三步骤，其中通过将在第一步骤中创建的第一依赖语法语法树与在第二步骤中创建的第二依赖语法语法树组合来创建第三依赖语法语法树; 第四步，其中通过使用和应用识别用于维护来自第三依赖语法语法树的每个节点感兴趣的节点的“REMAIN”操作的人造智能模型来创建修改的第四依赖语法语法树，“DELETE “用于删除感兴趣的节点及其所有子节点的操作，或者用于删除感兴趣节点的”CHILD_UP“操作，然后将其子节点放置在其中; 以及第五步骤，其中从修改的第四依赖语法语法树创建摘要。