SPES:基于谓词选择率估计的SPARQL查询优化方案

SPES: SPARQL Query Optimization Based on Predicate Selectivity Estimation

面对海量的RDF数据查询及查询语句本身复杂性的增长,提高SPARQL查询效率成为语义网研究的热点.在实际的大量查询请求中,SPARQL语句都是带有多谓词的查询.多谓词查询的复用可以减少重复查询或者缩减查询范围,从而有效的提高SPARQL查询性能.因此,提出一种基于谓词选择率估计的SPARQL查询优化方案.本文基于查询日志构建直方图,以统计历史查询语句的SPE(SPARQL谓词表达式),并利用构建的直方图指导缓存.本文提出了缓存的构建以及维护方案,其中谓词选择率估计在构建缓存时起着重要的作用.实验结果分析表明,本文提出的基于查询日志构建谓词直方图的方法可以有效提高查询效率,从而达到SPARQL查询优化的目的.本文将此方案命名为SPES.

Faced with the massive and complexity of RDF data query,the efficiency of SPARQL query has become a hot topic in the research of semantic web.Actually,lots of SPARQL statements have multiple predicates.Multiplexing of multi predicate queries can reduce the number of repeated queries and the range of query,and improve the performance of SPARQL query effectively.Therefore,this paper proposes a SPARQL query optimization scheme based on predicate selection rate estimation.In this paper,we construct a histogram,based on the query log,to gather statistics ofthe SPE （SPARQL predicate expression） of the historical query statement,and the histogram can be used to guide the work of the cache.This paper puts forward construction and maintenance scheme of the cache,in which the predicate selection rate plays an important role.Experimental results show that the method can improve the query efficiency,so as to achieve the purpose of SPARQL query optimization.Scheme of the paper called SPES.