计算机与现代化 ›› 2013, Vol. 1 ›› Issue (3): 34-37,4.doi:

• 算法分析与设计 • 上一篇    下一篇

基于本体的自动文摘方法研究与实现

朱 平1,费本华1,范少辉1,王 虎2   

  1. 1.国际竹藤网络中心,北京100102;2.上海大学计算机工程与科学学院,上海200072
  • 收稿日期:2012-11-13 修回日期:1900-01-01 出版日期:2013-04-03 发布日期:2013-04-03

Research and Implementation of Automatic Summarization Method Based on Ontology

ZHU Ping1, FEI Ben-hua1, FAN Shao-hui1, WANG Hu2   

  1. 1. International Centre for Bamboo Rattan, Beijing 100102, China; 2. School of Computer Engineering and Science, Shanghai University, Shanghai 200072, China
  • Received:2012-11-13 Revised:1900-01-01 Online:2013-04-03 Published:2013-04-03

摘要: 传统自动文摘方法并没有在语义层面上处理问题,所以近年来涌现出基于语义的理解分析法。本文提出一种基于本体的自动文摘方法,该方法采用领域本体将词特征映射为概念特征,然后调整特征词和文本语句权值,最后提取出合适的句子组合成文摘。该方法在统计的基础上加入语义分析,从而能够更精确地得到文本的主要内容,更方便阅读。在对竹藤领域的文本进行自动文摘实验后证明,该方法相比于传统自动文摘方法在文摘内容的准确度上有了较大的提高。

关键词: 自动文摘, 本体, 概念映射

Abstract: The traditional automatic summarization methods cannot work in the semantic level, a automatic summarization method based on ontology is proposed. Firstly, word features are mapped to concept features according to domain ontology, and then concept features weights and sentences weights will be adjusted, the last the important sentences can be selected and composed the summarization. The semantic analysis is added to the automatic summarization method which based on statistics. By using this method, the main content of the text can be more exact and easier to read. The automatic summarization experiments in the domain of bamboo and rattan prove that compared to traditional automatic summarization method, the accuracy of summarization has been greatly improved based on this method.

Key words: automatic summarization, ontology, concept map