计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (20): 177-181.

• 数据库、信号与信息处理 • 上一篇    下一篇

一种XML的树匹配改进方法

于亚君1,2,姜  瑛1,2,3   

  1. 1.昆明理工大学 信息工程与自动化学院,昆明 650500
    2.云南省计算机技术应用重点实验室,昆明 650500
    3.昆明理工大学 质量发展研究院,昆明 650093
  • 出版日期:2012-07-11 发布日期:2012-07-10

Improved method of tree matching using XML

YU Yajun1,2, JIANG Ying1,2,3   

  1. 1.Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China
    2.Yunnan Key Lab of Computer Technology Application, Kunming 650500, China
    3.Quality Development Institute, Kunming University of Science and Technology, Kunming 650093, China
  • Online:2012-07-11 Published:2012-07-10

摘要: 基于XML树的匹配已被广泛应用于数据挖掘、自然语言自处理、图像检索等领域。通过分析现有的基于XML树的匹配度计算方法,发现存在对计算的前期要求(如权值分割)太过严格、匹配度结果存在误差等问题,影响了匹配的精度和效率。基于XML的内容约束和结构约束,综合结点相似度和层次相似度,提出一种结构相似度计算公式,改进了匹配计算结果的准确度,并通过实验验证了公式的有效性。

关键词: 可扩展标示语言(XML), 树匹配, 相似度计算

Abstract: Tree matching based on XML has been widely used in many domains, including data mining, natural language processing, image retrieval and etc. The existing tree matching calculation methods based on XML are studied. There are some problems that founded in the existing method of matching calculation, such as strict requirement before calculation(weight division), errors in the result of matching, which will affect the matching accuracy and efficiency. In order to resolve the existing problems in formula and achieve better matching results, this paper computes the node similarity and the level similarity based on the constraints of content and structural of XML. A formula of the structural similarity is presented, which can improve the accuracy of matching results. The validity of the formula is proved through experiments.

Key words: Extensive Makeup Language(XML), tree matching, similarity calculation