首页> 外国专利> SYSTEM FOR IDENTIFYING ARTICLES AND METHOD TO DETERMINE THE AUTHENTICITY OF PROPOSITIONS REFLECTING THE RELIABILITY OF MEDIA

SYSTEM FOR IDENTIFYING ARTICLES AND METHOD TO DETERMINE THE AUTHENTICITY OF PROPOSITIONS REFLECTING THE RELIABILITY OF MEDIA

机译:用于确定文章和方法以确定反映媒体可靠性的命题真实性的系统

摘要

According to an embodiment of the present invention, there is provided a method for learning a machine learning model for article discrimination performed by an article discrimination system reflecting the reliability of the media, the method comprising: receiving a target proposition for authenticity determination; performing web crawling based on the received target proposition to search for a related article, and calculating a similarity between the target proposition and a sentence included in the searched related article; determining a sentence as true, false, or neutral according to the calculated similarity of the sentence and setting a discriminant value; And article discrimination machine learning comprising the step of performing learning by inputting the information of the media that provided the related articles including the sentences, the discriminant values set in the sentences, and the sentences, as learning data, into the machine learning model for which the similarity was calculated. We want to provide a method for learning the model.
机译:根据本发明的实施例,提供了一种用于学习由反映介质可靠性的物品辨别系统执行的物品辨别机器学习模型的方法,该方法包括:接收真实性确定的目标主张;基于所接收的目标命题进行网络爬网以搜索相关文章,并计算目标命题与搜索相关文章中包括的句子之间的相似性;根据句子的计算相似性确定一个真实,假或中立的句子,并设定判别值;文章辨别机学习包括通过输入包括句子的相关文章的媒体信息来执行学习的步骤,句子中设置的判别值以及句子作为学习数据,进入机器学习模型计算相似性。我们希望提供一种学习模型的方法。

著录项

  • 公开/公告号KR102326972B1

    专利类型

  • 公开/公告日2021-11-16

    原文格式PDF

  • 申请/专利权人 고려대학교 산학협력단;

    申请/专利号KR20190098775

  • 发明设计人 정창성;서영경;

    申请日2019-08-13

  • 分类号G06F40/20;G06N3/08;

  • 国家 KR

  • 入库时间 2022-08-24 22:30:01

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号