电子学报 ›› 2015, Vol. 43 ›› Issue (7): 1320-1328.DOI: 10.3969/j.issn.0372-2112.2015.07.011

• 学术论文 • 上一篇    下一篇

基于语义度量的RDF图近似查询

章登义, 吴文李, 欧阳黜霏   

  1. 武汉大学计算机学院, 湖北武汉 430072
  • 收稿日期:2014-12-01 修回日期:2015-01-29 出版日期:2015-07-25 发布日期:2015-03-24
  • 作者简介:章登义 男,1955年2月出生于湖北省荆州市.现为武汉大学计算机学院教授、博士生导师.主要研究方向为安防、嵌入式和数据库. E-mail:dyzhang@whu.edu.cn;吴文李 女,1986年12月出生于广东省湛江市.现为武汉大学博士研究生.主要研究方向为语义网数据挖掘. E-mail:huihuigou@whu.edu.cn

Approximating Query with Semantic-Based Measure on RDF Graphs

ZHANG Deng-yi, WU Wen-li, OUYANG Chu-fei   

  1. School of Computer, Wuhan University, Wuhan, Hubei 430072, China
  • Received:2014-12-01 Revised:2015-01-29 Online:2015-07-25 Published:2015-03-24

摘要:

近似查询是图数据库资源管理的操作之一.已有工作主要基于距离来度量查询语句与图的近似值,忽略了两者之间的语义近似性.对于语义图的近似查询,忽略图与查询的语义近似将难以有效完成查询.针对该问题,本文在考虑语义近似的基础上为RDF(Resource Description Framework,资源描述框架)图的近似查询提出基于语义距离的度量方法.同时,为提高查询效率,本文提出语义结构剪枝策略.最后,我们构造查询框架以实现查询的响应过程,并在该框架下设计实验以评价本文方法.实验表明,本文方法可高效执行RDF近似查询并有效返回top-k结果集.

关键词: RDF图, 近似查询, 图数据库, 查询处理

Abstract:

Approximating query is one of the operations for resource management in graph database.Existing works mainly based on the distance similarity to measure the query and graphs without considering the semantic similarity.For the approximating query on semantic graphs,disregarding the semantic similarity may fail the query.In this paper,we propose a semantic-based measure for approximating query on RDF graphs,considering the semantic similarity.In the meanwhile,we specify a semantic structural pruning strategy to ameliorate the efficiency of query process.Finally,we construct the query framework to answer approximating query,and design experiments to test our methods under this framework.Results show that the approaches in this paper can efficiently execute approximating query on RDF graphs and effectively return top-k results.

Key words: RDF graph, approximating query, graph database, query processing

中图分类号: