模糊XML关键字近似查询方法研究  

Research of Approximate Keyword Query on Fuzzy XML Documents

作　　者：李婷[1] 程海涛[1] 

机构地区：[1]东北大学计算机科学与工程学院,沈阳110819

出　　处：《计算机科学》2017年第9期216-221,226,共7页Computer Science

摘　　要：在精确XML文档上的关键字查询方法的研究大多是基于LCA语义或者其变种语义(SLCA,ELCA等)开展的,将包含所有关键字的最紧致XML子树片段作为查询结果返回。但是这些基于LCA语义产生的查询结果中通常包含了大量的冗余信息,现实世界中存在着大量的不确定和模糊信息,因而如何从模糊XML文档中搜索到高质量的关键字查询结果是一个需要研究的问题。针对模糊XML文档上的关键字近似查询方法进行研究,通过引入最小连接树(MCT)的概念,提出在模糊XML文档上关键字查询的所有GDMCTs问题,并给出解决这一问题的基于栈的算法All fuzzy GDMCTs,该算法可以得到满足用户指定的子树大小阈值和可能性阈值条件的所有GDMCTs结果。实验表明,该算法在模糊XML文档上能够得到较高质量的关键字查询结果。The study of keyword queries on crisp XML documents is carried out mainly based on the LCA semantics or its variant semantics （SI.CA,EI.CA）,and the most compact XML subtrees containing all keywords are returned as the query results. However,the generated results based on the LCA semantics always contain a lot of redundant informa- tion,and there are a lot of uncertainty and fuzzy information exist in the real world. How to search the high quality re- suits of keyword queries on fuzzy XML documents is an issue need to be studied. Aiming at investigating the method of approximate keyword query on fuzzy XMI. documents, firstly the concept of minimum connecting tree was introduced, All GDMCTs problem of keyword queries on fuzzy XML documents was proposed, and a stack based algorithm All fuzzy GDMCTs was given to solve the problem. The algorithm can get all the GDMCTs results satisfying the given sub tree size threshold and possibility threshold. Experimental results show that the algorithm can get the high quality re- suits of keyword queries on fuzzy XML documents.

关 键 词：XML 关键字 近似查询 模糊 可能性 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]