METHOD APPARATUS AND SYSTEM FOR DETERMINING SIMILARITY OF PATENT DOCUMENTS USING CLUSTERING

Provided is a method of determining the similarity of a patent document. The method of determining the similarity of a patent document includes: a step wherein a server obtains a similar patent document by inputting a target patent document into a first artificial intelligence model; a step wherein...

Full description

Saved in:
Bibliographic Details
Main Authors KIM DOUN, PARK SANGJUN
Format Patent
LanguageEnglish
Korean
Published 12.04.2021
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Provided is a method of determining the similarity of a patent document. The method of determining the similarity of a patent document includes: a step wherein a server obtains a similar patent document by inputting a target patent document into a first artificial intelligence model; a step wherein the server obtains a plurality of sentences included in the target patent document, and obtains a plurality of sentences included in the similar patent document; a step wherein the server obtains a first sentence of the plurality of sentences included in the target patent document, and obtains a second sentence of the plurality of sentences included in the similar patent document; and a step wherein the server inputs the first and second sentences into a second artificial intelligence model to obtain an evaluation result about the first and second sentences. Therefore, the present invention is capable of guaranteeing a stable result with a small amount of time invested. 특허문서의 유사도 판단 방법이 제공된다. 상기 특허문서의 유사도 판단 방법은, 서버가, 대상특허문서를 제1 인공지능 모델에 입력하여 유사특허문서를 획득하는 단계; 상기 서버가, 상기 대상특허문서에 포함된 복수의 문장을 획득하고, 상기 유사특허문서에 포함된 복수의 문장을 획득하는 단계; 상기 서버가, 상기 대상특허문서에 포함된 복수의 문장 중 제1 문장을 획득하고, 상기 유사특허문서에 포함된 복수의 문장 중 제2 문장을 획득하는 단계; 상기 서버가, 상기 제1 문장 및 상기 제2 문장을 제2 인공지능 모델에 입력하여, 상기 제1 문장과 상기 제2 문장에 대한 평가 결과를 획득하는 단계; 를 포함한다.
Bibliography:Application Number: KR20200034425