计算机与数字工程2017，Vol.45Issue(12)：2474-2478,5.DOI:10.3969/j.issn.1672-9722.2017.12.031
基于超图的多文档新闻关键词抽取
News Keyword Extraction in Multi-documents based on Hypergraph
范泽泉 1赖华2
作者信息
- 1. 昆明理工大学信息工程与自动化学院 昆明 650500
- 2. 昆明理工大学信息处理重点实验室 昆明 650500
- 折叠
摘要
Abstract
As an important carrier of network information dissemination,the essence of the news is a continuous process aimed at reaching the truth.As time goes by,there will be a large number of different web pages for the same news event.How to quickly and accurately extract the key information of these news has become an increasingly important issue. Keywords as a brief summary of the article content,which can be used to quickly understand the news events,thus saving a lot of time,so keyword ex?traction technology is considered to be the key to solve this problem.Based on the analysis of the characteristics of news web pages, this paper proposes a new method for extracting multi-document keywords based on hypergraph model.This method takes words as nodes and news web pages as hyperedges,considering the factors of web-trust and the time of news,a hypergraph model of multi-news documents is established. Finally,the keywords are extracted by hypergraph ranking algorithm. The experimental re?sults verify the accuracy of the method.关键词
多文档超图模型/超图排序/随机游走/关键词抽取/网页信任度/时间因素Key words
multi-document hypergraph model/hypergraph ranking/random walk/keywords extraction/web-trust/time factor分类
信息技术与安全科学引用本文复制引用
范泽泉,赖华..基于超图的多文档新闻关键词抽取[J].计算机与数字工程,2017,45(12):2474-2478,5.
