基于树结构RDF闭包生成算法研究  

Generating Closure of RDF Source Based on Tree Structure

作　　者：尹康银[1] 宋自林[1] 乔可春[1] 陈博[1] 

机构地区：[1]解放军理工大学指挥自动化学院,南京210007

出　　处：《系统仿真学报》2008年第4期1072-1075,1079,共5页Journal of System Simulation

基　　金：国防基金(51406020105JB8103)

摘　　要：RDF闭包是提高RDF数据查询效率的一种有效途经,目前大部分闭包生成算法依据推理规则间的触发实现推理,推理规则多次重复使用,闭包生成效率比较低。利用树的层次结构思想,根据RDF(S)语义推理规则的特点,提出基于树的闭包生成算法,将RDF(S)推理规则数据结构化,构造两棵分别对应RDF属性和概念的属性树和概念树,树的节点存储RDF三元组。然后根据映射机制将存储在树节点中的三元组映射为RDF闭包。仿真显示此方法有效地提高了闭包生成效率。RDF closure is an effective way to improve the RDF query efficiency. Existing closure algorithms almost perform reasoning according to the triggers among the inference rules which are used repeatedly, and thus the efficiency of closure generation is relatively lower. By virtue of the characteristics of the inference rules of RDF （S） semantics, a closure algorithm was proposed inspired with the idea of the hierarchy structure of trees. Property-tree and concept-tree corresponding to property and concept were respectively built, and the inference rules of RDF （S） semantics were represented with tree structure. RDF triples were stored in the nodes of property-tree and concept-tree. Then the RDF triples were mapped to RDF closure according to the mapping mechanism. Simulation demonstrates that this algorithm greatly improves the efficiency of closure generation.

关 键 词：本体论 RDF闭包 推理规则 属性-概念树 映射 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]