一种基于XML的统一构件匹配技术  

A United Component Matching Technology Based on XML

作　　者：曾一[1] 刘元勇[1] 郭永林[1] 

机构地区：[1]重庆大学计算机学院,重庆400044

出　　处：《计算机科学》2007年第3期279-282,共4页Computer Science

摘　　要：构件检索和匹配是CBSD和软件复用的关键技术。目前构件库所采用的关键字、刻面、规约、行为等匹配技术都存在两个问题:1)构件表示方法各异,理解和共享构件困难;2)每个查询匹配算法侧重点不同,关键字和刻面匹配侧重构件的语义,而规约匹配侧重构件的结构,造成单独使用一种匹配算法或多或少产生一些冗余或无关构件。本文提出一个XML统一构件描述模型,该模型利用XML模式对构件进行层次化的组织,采用一种最大权匹配算法,从语法和语义两方面计算两个构件间的匹配度,有效地解决了上述问题。通过实验证明,该技术在保证构件查全率的基础上有效地提高了查准率。Component retrieving and matching are one of key technology of CBSD and software reuse. Component matching technology used now in component repository has two main problems. One is that different expressions of components leads to the difficult understanding and sharing of components, the other is that different arithmetic of matching emphasizes particularly on components which causes some redundancy and irrelated components. An uniform component description model which makes use of XML Schema to organize components hierarchically has been brought forward in the article and the author also introduces a maximal weight matching method for calculating matching degree of components in semantic and syntax. It is testified that it can not only help users to lookup the component, but also improve the precision of the searching.

关 键 词：软件复用 XML 构件检索 构件匹配 

分 类 号：TP311.132[自动化与计算机技术—计算机软件与理论]