METHOD APPARAUTS AND SYSTEM FOR NAMED ENTITY LINKING AND COMPUTER PROGRAM THEREOF

The present invention relates to a method, apparatus, system and computer program for named entity linking. More particularly, in a method for performing named entity linking for an ambiguous word included in a text, disclosed is the method for named entity linking, including: a word-vector conversi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SOHN, DAE NEUNG, KANG, IN HO, SEO, DEA RYONG, CHANG, SUNG EUN, JANG, YOUNG WHAN, KIM, HYO SEUNG, LEE, DONG JU, KODAMA CHIAKI, CHUNG, YOU JIN, SON, HOON SEOK, BAIK, SONG IY
Format: Patent
Sprache:eng ; kor
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The present invention relates to a method, apparatus, system and computer program for named entity linking. More particularly, in a method for performing named entity linking for an ambiguous word included in a text, disclosed is the method for named entity linking, including: a word-vector conversion step of converting a plurality of words including the ambiguous word in the text into a plurality of corresponding vectors; a neural network input step of inputting the plurality of vectors to a deep neural network including a hidden layer; and a named entity linking step of performing the named entity linking for the ambiguous word using an output value of the deep neural network, by the apparatus for named entity linking. Accordingly, the present invention can perform accurate named entity linking for the ambiguous word located in the text. 본 발명은 개체명 연결 방법, 장치, 시스템 및 컴퓨터 프로그램에 관한 것으로서, 보다 구체적으로는 텍스트에 포함되는 다의어에 대하여 개체명 연결을 수행하는 방법에 있어서, 개체명 연결 장치가, 상기 텍스트에서 상기 다의어를 포함하는 복수의 단어를 각 대응하는 복수의 벡터로 변환하는 단어-벡터 변환 단계; 상기 복수의 벡터를 은닉층을 구비하는 심층신경망에 입력하는 신경망 입력 단계; 및 상기 심층신경망의 출력값을 이용하여 상기 다의어에 대한 개체명 연결을 수행하는 개체명 연결 단계;를 포함하는 것을 특징으로 하는 개체명 연결 방법을 개시한다.