【作者】 王桐； 刘大昕； 田迪； 孙伟； 张万松；

【Author】 Wang Tong~1,Liu Daxin~1,Tian Di~2,Sun Wei~1,and Zhang Wansong~1 1(College of Computer Science and Technology,Harbin Engineering University,Harbin 150001) 2(Travel Sky Airport Technology Co.,Harbin 150090)

【机构】 哈尔滨工程大学计算机科学与技术学院； 黑龙江航信空港网络有限公司；

【摘要】 XML已成为网上数据交换和存储的标准,然而现有的XML文档模型存在很多问题.综合了数据挖掘和信息检索两方面的技术,对传统的向量空间模型加以扩展,提出一种新的基于语义和支持度的XML向量空间模型并给出其生成算法,该模型抽取文档集的频繁路径作为特征,然后利用XML本身的语义特性,对标签之间作ontology判断.最后,基于该模型提出一种XML近似查询算法,将文档矩阵从改进的VSM空间映射到潜在语义空间,并在转换后的空间获得近似查询结果并排序.对提出的方法进行理论分析和实验验证,得到满意的结果.更多还原

【Abstract】 XML has become the de facto standard of data interchange and storage,however,there are some defects of the current XML model.Considering both data mining and information retrieval techniques,the vector space model is extended and a novel XML model based on ontology and supports is proposed.Also given is the constructing algorithm.The model at first extracts the frequent sequences from the document collection as the features,and then judges the semantic features between the tags of XML documents using ontology.Furthermore,an approximate querying method(XAQ) is presented based on the XML model. This algorithm maps the documents set from the extended VSM to the LSI space,matches the queries and ranks the returned results.The theoretical analysis is given to the proposed method and the experiment shows the satisfactory results.更多还原