细粒度语义网检索  

Fine-grained semantic web retrieval

作　　者：吴刚[1] 唐杰[1] 李涓子[1] 王克宏[1] 

机构地区：[1]清华大学计算机科学与技术系,北京100084

出　　处：《清华大学学报（自然科学版）》2005年第S1期1865-1872,共8页Journal of Tsinghua University(Science and Technology)

基　　金：国家自然科学基金资助项目(60443002)

摘　　要：语义网的有向标记图数据模型决定其在内容检索方面与纯文本、超文本或半结构化文档检索存在较大差异。现有检索模型和相似度计算方法不能完全满足对语义网的检索和评价。该文以资源这种较细粒度作为检索单元。Semantic web uses directed labeled graph as its fundamental data model which is different from plain text, hyper-text, and semi-structured document. Obviously, the traditional information retrieval model and similarity computation can not be directly applied to the semantic web retrieval. In this paper, resource is used as a basic unit for indexing and retrieving, which results in a fine-grained retrieval model. The model considers both the distribution of term frequency and the semantic structure according...

关 键 词：在资源内容中的频率和结构分布 提出一个细粒 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]