细粒度语义网检索 | |
吴刚 ; 唐杰 ; 李涓子 ; 王克宏 ; WU Gang ; TANG Jie ; LI Juanzi ; WANG Kehong | |
2010-06-09 ; 2010-06-09 | |
关键词 | 在资源内容中的频率和结构分布 提出一个细粒 computer software semantic web information retrieval similarity computation inverted index TP391.3 |
其他题名 | Fine-grained semantic web retrieval |
中文摘要 | 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,考虑; 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 to characteristics of the semantic web. The traditional similarity computation and the inverted index structure are modified with respect to the new model and implemented in SWARMS semantic web aiding rich mining system, a semantic web mining system. Experiments on two data sets, SourceForge and DBLP, show that our method has the ability to combine the evaluation on content and structure queries to the semantic web. At the same recall level, this combination doubles the precision value.; 国家自然科学基金资助项目(60443002) |
语种 | 中文 ; 中文 |
内容类型 | 期刊论文 |
源URL | [http://hdl.handle.net/123456789/56003] |
专题 | 清华大学 |
推荐引用方式 GB/T 7714 | 吴刚,唐杰,李涓子,等. 细粒度语义网检索[J],2010, 2010. |
APA | 吴刚.,唐杰.,李涓子.,王克宏.,WU Gang.,...&WANG Kehong.(2010).细粒度语义网检索.. |
MLA | 吴刚,et al."细粒度语义网检索".(2010). |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论