Optimizing SPARQL queries in Linked Open Data using Heuristic approach

Gouri D. Potdar, Rachel Dhanaraj

Abstract


As today’s world tends to rely more and more on search engines for quenching its thirst for information, search engines, today, are expected to be faster, more accurate, more intelligent and more powerful so as to reach a wide pool of information resources. Data sources like CKAN1, DBpedia2, GeoNames3, FOAF4 which collectively form Linked Open Data (LOD) have gained importance in this quest for better search engines. SPARQL is the w3c recommended query language which is used to extract data from LOD sets. SPARQL queries typically contain more joins than equivalent relational plans, and hence lead to a large join order search space. Consequently, query optimization in RDF Stores is a challenge. The dynamic nature of LOD prevents the application of the cost based approach which requires statistics. Moreover, the relevant correlations cannot be identified beforehand. Hence, using good heuristics for SPAQRL query optimization is an advantage.


Keywords—Linked Open Data, Heuristics, Query Optimization, SPARQL, RDF


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v5i4.2081

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science