【作者】 姚俊杰； 崔斌；

【机构】 南开大学计算机科学与技术系； 北京大学计算机科学系；

【摘要】 <正>1引言随着XML数据的持续增加,有效处理XML数据并为决策提供信息支持变得日益重要。XML允许半结构化和层次化表示,对XML数据的挖掘不同于传统结构化数据和文本数据。XML的挖掘研更多还原

【Abstract】 With the standardization of XML as an information exchange language over the Web,a huge amount of information is formatted in XML documents.The semi-structural nature of XML means that,the structural information is an important clue as to the meaning of XML documents.The structural clustering result plays an important role in many web information retrieval and XML manipulation tasks.As to the short coming of a path-based XML document clustering algorithm on robust and scalability,we have the algorithm improved by replacing similarity measure,weighting clustering features and introducing new algorithms.The experiments show that our improved algorithm outperforms the old one and is scalable well.更多还原