XML数据相似度研究  

A Recursive Method to Compute Similarity of XML Documents

作　　者：张丙奇[1] 白硕[1] 赵章界[1] 

机构地区：[1]中国科学院计算技术研究所,北京100080

出　　处：《计算机工程》2005年第11期25-27,126,共4页Computer Engineering

摘　　要：XML数据的大量出现为信息检索、数据挖掘、智能信息处理提供了机遇和挑战,而相似度计算是XML文档检索、挖掘和深层次智能处理的基础,对相似度计算进行研究具有非常重要的意义。在对XML数据特征进行深入分析的基础上,提出了一种递归相似度计算方法,实验结果表明该方法具有较好的效果。The quantity of XML data shared in the World Wide Web is increasing quickly and it offers both challenges and opportunities in information retrieval, data mining and intelligent information processing. As the basis of information retrieval, mining, and processing, accurate determination of similarity between XML documents is important and valuable. This paper provides a novel recursive method to get the similarity between XML documents according to the semantic and structural features of XML. It also provides experiments to show the comparison of the method against traditional methods and the results prove the method is highly effective.

关 键 词：XML 相似度 语义 递归算法 VSM模型 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]