摘要
为了更好地为消费者提供具有参考价值的在线评价,本论文基于对网上评论内容的分析,借助LDA主题模型挖掘出评论内容中所隐藏的主题信息,并与标准训练语料的主题信息进行对比,计算它们之间的信息熵,使用计算出来的信息熵来表示评论内容偏离标准语料库的程度,从而得到评论内容的有用程度。
In order to better provide consumers with online evaluations of reference value,this thesis is based on the analysis of on line comments,use the LDA theme model to dig out the hidden subject information in the comments,and compare it to the subject matter of the standard training corpus,calculate the information entropy between them,use the calculated information entropy to in dicate the degree to which the content of the comment deviates from the standard Corpus,to obtain the usefulness of the content of the comments.
作者
陈雪晶
程锐
CHEN Xue-jing;CHENG rui(Yangtze University,Jingzhou 434023,China)
出处
《电脑知识与技术》
2019年第9Z期266-268,共3页
Computer Knowledge and Technology
