【作者】 王小锋； 张新； 谢敏； 孟小峰； 周军锋；

【Author】 Wang Xiaofeng~1,Zhang Xin~1,Xie Min~1,Meng Xiaofeng~1,and Zhou Junfeng~(1,2) 1(School of Information,Renmin University of China,Beijing 100872) 2(Department of Computer Science and Technology,Yanshan University,Qinhuangdao 066004)

【机构】 中国人民大学信息学院； 燕山大学计算机科学与技术系；

【摘要】 XML数据流上的XPath & XQuery查询处理是目前研究者关注的热点问题,但由于XPath &XQuery查询语言相对复杂,在不知道模式信息的前提下,用户很难通过已有的查询接口得到自己感兴趣的数据片断,因此如何在数据流模型上根据XML数据的特点为用户提供最友好的查询接口就成为一个亟待解决的问题.针对这个问题,创新地提出了在XML数据流上做关键字查询的问题,给出了最小相关连通子树(SRCT)的概念用于处理返回的结果,并设计了一种新的基于栈的Lookup算法,可以有效解决在XML数据流上进行关键字查询的问题,最后通过实验从不同角度对Lookup算法的各项性能指标进行了验证.更多还原

【Abstract】 XPath & XQuery query processing on XML streams has become the hot topic and attracted many researchers’ attention.However,due to the relatively complex query language of XPath & XQuery, it is hard for users to get interesting data fragments via existing query interface without the prior knowledge of the schema information.As a result,it is becoming an impending issue to address the problem to provide uses with the most friendly query interface in XML streams according to the features of XML data.In this paper,the problem of keyword search on XML streams is proposed for the first time and the concept of smallest related connected subtree(SRCT) is presented,In addition,a novel stacked-based lookup algorithm is designed,which can efficiently solve the problem of keyword search on XML streams. Sufficient experiments show that the lookup algorithm is efficient in practice in various aspects.更多还原