基于术语编码和树匹配的构件检索方法研究  

Study on Approach of Term Encoding and Tree Matching Based on Component Retrieval

作　　者：渠成建[1] 陈立潮[1] 潘理虎[1] 张英俊[1] 谢斌红[1] 

机构地区：[1]太原科技大学计算机科学与技术学院,太原030024

出　　处：《电脑开发与应用》2012年第9期39-42,共4页Computer Development & Applications

基　　金：山西省自然科学基金(2009011022-1);太原科技大学研究生创新项目

摘　　要：在研究和分析已有编码和树匹配检索方法的基础上,给出了一种基于术语编码和树匹配的改进方法。通过对构件库中的术语进行编码来建立索引,依据树匹配思想计算构件的匹配代价和匹配度,按照匹配代价排序检索结果,最后根据匹配度要求提取并输出符合要求的构件。实验表明,该方法在保证较高查全率的基础上,能够有效提高构件的查准率。Effective component retrieval is a critical factor of software reuse. Based on the iaeeted classification and the theory of tree matching, an improved component retrieval method of term encoding and tree matching is proposed. An index is created by encoding all the terms of the component library in this method, and the matching cost and matching degree are calculated based on the tree matching. After calculating, the components are sorted according to the matching cost and picked up to the result set while the requirement of the matching degree is met. The experiment shows that this approach can improve the precision effectively while keeping high recall ratio.

关 键 词：术语编码 术语索引 匹配代价 匹配度 构件检索 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]