计算机技术与发展2016，Vol.26Issue(4)：110-113,4.DOI:10.3969/j.issn.1673-629X.2016.04.024
基于分级策略和聚类索引树的构件检索方法
A Component Retrieval Method Based on Classified Policy and Cluster Index Tree
摘要
Abstract
In the component representation method based on faceted classification,the term-space needs human to build and maintain so as to have strong subjectivity. Therefore,a component representation method combined faceted classification with full-text retrieval has been adopted in this paper. Meanwhile,a component retrieval method based on classified policy and cluster index tree has been proposed. In this method,the component cluster index tree is built by use of a component clustering algorithm based on semantic similarity and opti-mization,and the reasonable weight factors are introduced for each facet. On the foundation of the real component library,the experiment shows that by the comparison of the component retrieval method without weigh factors, the component retrieval method proposed has higher precision ratio and recall ratio,and to some extent,achieves component semantic retrieval.关键词
刻面分类/聚类分析/语义分析/索引树/分级策略Key words
faceted classification/cluster analysis/semantic analysis/index tree/classified policy分类
信息技术与安全科学引用本文复制引用
王文霞..基于分级策略和聚类索引树的构件检索方法[J].计算机技术与发展,2016,26(4):110-113,4.基金项目
国家自然科学基金资助项目(11241005) （11241005）
山西省高等学校教学改革研究项目(J2012098) （J2012098）
运城学院教学改革研究项目(JG201418) （JG201418）
