CAJ | 학술논문

网络舆情分析工作中的一项关键任务是识别新闻之间的多次转载关系，进而确定目标事件舆情在互联网上的分布状况和传播范围。基于网页主题内容抽取、新闻转载关系初筛，以及基于核函数的相似度计算等环节，判别新闻报道之间的转载关系，并对转载识别效果进行检验，构建了基于网页转载关系判别的网络舆情传播态势分析模型，有助于对敏感事件的网络舆情进行针对性的引导和控制。
망락여정분석공작중적일항관건임무시식별신문지간적다차전재관계，진이학정목표사건여정재호련망상적분포상황화전파범위。기우망혈주제내용추취、신문전재관계초사，이급기우핵함수적상사도계산등배절，판별신문보도지간적전재관계，병대전재식별효과진행검험，구건료기우망혈전재관계판별적망락여정전파태세분석모형，유조우대민감사건적망락여정진행침대성적인도화공제。
Identifying the near-duplicate web pages helps to confirm the spread scope of public opinion on the Internet. In this paper we studied how to identify the reprinted relations between news content. Firstly, we extract the news content from the web pages;secondly, we find the candidates of reprinted web pages;then we use the kernel function to measure the similarity between the news content, and find out the reproduced relations. The experiment result shows that the model proposed in this paper can effectively identify the reproduced rela-tions between web pages. The model is helpful for prevention and control of the negative Internet public opinion.