基于局部话题句群的事件相关多文档摘要研究

多文档自动文摘研究的目的是给用户提供简洁全面的文档信息并提高用户获取信息的效率。在进行局部话题确定时，通常是利用聚类分析的方法把相似的文本单元聚成一个局部话题。该文提出了一种针对新闻事件的多文档摘要生成方法，其特色在于：在提取基本新闻要素和扩展新闻要素的基础上分别形成了基本局部话题句群（BPTSG）和扩展局部话题句群（EPTSG），这样可以在尽可能全面地覆盖多个话题的同时缩减自身的冗余。此外，文中还提出了一种基于事件时间和句子位置信息的文摘句排序方法。实验结果验证了该文所提的方法是有效的，与基于聚类的自动文摘系统相比较，该系统生成的摘要质量有显著提高。

Study on Event Relevant Multi-Document Summarization Based on Partial-Topic-Sentence Group

The aim of multi-document automatic summarization is to provide people a concise and full-information document and help people get information effectively.When partial topic is identified,clustering method is usually used to cluster similar document units into a partial topic.A multi-document summarization method focused by events is proposed in this paper,the specificity of the method is that the basic-partial-topic-sentence group(BPTSG) and the extended-partial-topicsentence group(EPTSG) are generated based on the extraction of the basic news factors and the extended news factors,which can both cover as many themes as possible and reduce its redundancy at the same time.In addition,a method based on event time and the position information of sentence is proposed to sort the abstract sentences.Experimental results indicate that the method proposed by this paper is effective,the proposed system is compared to the clustering-based system and it is shown that it improves summaries quality remarkably.