首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于互增强关系的自动文摘句子加权方法
引用本文:王志琪,王永成,刘传汉.基于互增强关系的自动文摘句子加权方法[J].上海交通大学学报,2007,41(8):1297-1300.
作者姓名:王志琪  王永成  刘传汉
作者单位:上海交通大学,计算机科学与工程系,上海,200240
基金项目:国家高技术研究发展计划(863计划)
摘    要:针对自动文摘处理,提出一种新颖的自动文摘句子加权方法.该方法采用基于互增强关系(MRP)的迭代算法模拟句子和词之间的循环加权关系,计算句子权重.实验结果表明,基于MRP的迭代算法收敛速度快;与传统的自动文摘方法和MS Word Summarizer相比,所提出的基于MRP的句子加权方法可以使自动文摘系统具有更好的性能.

关 键 词:自动文摘  互增强关系  迭代算法  句子加权
文章编号:1006-2467(2007)08-1297-04
修稿时间:2006-08-31

A Sentence Weighting Method for Automatic Text Summarization Based on MRP
WANG Zhi-qi,WANG Yong-cheng,LIU Chuan-han.A Sentence Weighting Method for Automatic Text Summarization Based on MRP[J].Journal of Shanghai Jiaotong University,2007,41(8):1297-1300.
Authors:WANG Zhi-qi  WANG Yong-cheng  LIU Chuan-han
Institution:Dept. of Computer Science and Eng. , Shanghai Jiaotong Univ. ,Shanghai 200240, China
Abstract:A novel method for text summarization was proposed.The mutual reinforcement principle(MRP) is used to simulate the cycle reinforcement relationship between terms and sentences.An iterative algorithm based on MRP is used to get the saliency scores of sentences.The evaluation shows that the iterative algorithm converges rapidly.Compared with the traditional summarizing method and MS Word Summarizer, the sentence weighting method based on MRP can improve the summarizing performance.
Keywords:automatic text summarization  mutual reinforcement principle(MRP)  iterative algorithm  sentence weighting
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号