Processing

Please wait...

Settings

Settings

Goto Application

1. WO2021112384 - NATURAL LANGUAGE REPRESENTATION METHOD AND COMPUTER PROGRAM STORED IN MEDIUM TO EXECUTE SAME

Publication Number WO/2021/112384
Publication Date 10.06.2021
International Application No. PCT/KR2020/013525
International Filing Date 06.10.2020
IPC
G06F 40/279 2020.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
20Natural language analysis
279Recognition of textual entities
G06F 40/205 2020.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
20Natural language analysis
205Parsing
G06F 40/268 2020.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
20Natural language analysis
268Morphological analysis
Applicants
  • 울산대학교 산학협력단 UNIVERSITY OF ULSAN FOUNDATION FOR INDUSTRY COOPERATION [KR]/[KR]
Inventors
  • 옥철영 OCK, Cheol Young
  • 이주상 LEE, Ju Sang
Agents
  • 김종선 KIM, Jong Sun
Priority Data
10-2019-016074205.12.2019KR
Publication Language Korean (KO)
Filing Language Korean (KO)
Designated States
Title
(EN) NATURAL LANGUAGE REPRESENTATION METHOD AND COMPUTER PROGRAM STORED IN MEDIUM TO EXECUTE SAME
(FR) PROCÉDÉ DE REPRÉSENTATION DE LANGAGE NATUREL ET PROGRAMME D'ORDINATEUR STOCKÉ DANS UN SUPPORT PERMETTANT DE L'EXÉCUTER
(KO) 자연어 표상 방법 및 이를 수행하기 위해 매체에 저장된 컴퓨터 프로그램
Abstract
(EN)
The present invention can rapidly and highly accurately represent and learn natural language by comprising: a natural language analysis step for dividing natural language into basic units; a part-of-speech information generation step for generating part-of-speech information corresponding to basic units; a meaning abstraction step for generating higher-level concept information comprising the basic units; and natural language representation step for processing the natural language using the part-of-speech information and higher-level concept information.
(FR)
La présente invention peut représenter et apprendre rapidement et avec précision un langage naturel car elle comprend : une étape d'analyse de langage naturel permettant de diviser un langage naturel en unités de base ; une étape de génération d'informations de partie de parole permettant de générer des informations de partie de parole correspondant à des unités de base ; une étape d'abstraction de signification permettant de générer des informations de concept de niveau supérieur comprenant les unités de base ; et une étape de représentation de langage naturel permettant de traiter la langue naturelle à l'aide des informations de partie de parole et des informations de concept de niveau supérieur.
(KO)
본 발명은 자연어를 기본 단위로 분리하는 자연어 분석 단계, 기본 단위에 품사 정보를 생성하는 품사 정보 생성 단계, 기본 단위를 포함하는 상위 개념 정보를 생성하는 의미 추상화 단계, 및 품사 정보 및 상위 개념 정보를 이용하여 자연어를 처리하는 자연어 표상 단계를 포함하여, 자연어 표상 및 학습을 빠르고 정확도 높게 수행할 수 있다.
Latest bibliographic data on file with the International Bureau