HLDA BASED SENTENCE SCORING FOR MULTI-DOCUMENT SUMMARY
首发时间:2013-10-29
Abstract:In recent years, the multi-document summary technology has gotten more and more attention in the field of natural language processing. However, the relationship between the topics and the level information are rarely considered, and sentence scoring is also a very important and difficult task in the multi-document summary process. The results of hLDA (hierarchical Latent Dirichlet Allocation) in the hierarchical topic modeling have been widely validated. Therefore this paper focused on the nodes in the hLDA model, researched the hLDA and semantic based sentence scoring method and presented seven algorithms to provide a strong basis for the multi-document summary.
keywords: natural language processing multi-document summary sentence scoring hLDA node
点击查看论文中文信息
基金:
论文图表:
引用
No.****
同行评议
共计0人参与
勘误表
基于hLDA主题模型的多文档摘要句子打分算法研究
评论
全部评论0/1000