| 基于局部主题判定与抽取的多文档文摘技术 |
| |
| 引用本文： | 秦兵,刘挺,李生.基于局部主题判定与抽取的多文档文摘技术[J].自动化学报,2004,30(6):905-910. |
| |
| 作者姓名： | 秦兵 刘挺 李生 |
| |
| 作者单位： | 1.哈尔滨工业大学计算机科学与技术学院,哈尔滨 |
| |
| 基金项目： | 国家自然科学基金(60203020)
国家“863”高科技项目基金(2001AA114041)资助~~ |
| |
| 摘 要： | 提出了一个通过对同一主题的多文档集合内局部主题的判定和抽取生成多文档文摘
的方法.首先在对多文档集合中句子依存分析和语义分析的基础上进行相似度计算,将相似
句子经过聚类形成多文档集合内不同的局部主题,然后进行每个局部主题中质心句的抽取和
排序,生成多文档文摘.该方法实现了文摘长度随文档内容自动确定,从而保证了文摘中包
含的信息的全面和简洁.最后文中还给出了多文档文摘的评价方法和实验结果,文摘的平均
精确率和平均压缩率分别为71.4%和25.2%.
|
| 关 键 词： | 多文档文摘 局部主题 聚类 |
| 收稿时间： | 2003-10-9 |
| 修稿时间： | 2003年10月9日 |
Multi-document Summarization Based on Local Topics Identification and Extraction |
| |
| QIN Bing,LIU Ting,LI Sheng.Multi-document Summarization Based on Local Topics Identification and Extraction[J].Acta Automatica Sinica,2004,30(6):905-910. |
| |
| Authors: | QIN Bing LIU Ting LI Sheng |
| |
| Affiliation: | 1.School of Computer Science and Tecchology,Harbin Institute of Technalogy,Harbin |
| |
| Abstract: | This paper describes a multi-document summarization method based on localtopics identification and extraction.The similarity of sentences is measured by analysis ofdependency and semantics.Local topics are found by sentence clustering.The centroidsentence is extracted from each local topic and is ordered to generate summarization.Thesize of summarization is determined according to content of multiple documents,as a result,the summarization becomes general and concise.Finally,the evaluation and experiment aregiven,the average precision of summarization and the average ratio of compressibility are71.4% and 25.2%,respectively. |
| |
| Keywords: | Multi-document summarization local topic clustering |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《自动化学报》浏览原始摘要信息 |
|
点击此处可从《自动化学报》下载全文 |
|
