This page provides access to the files for the paper "Efficient Inference-aware RDB2RDF Query Rewriting". These are the links:

• FAO testcases. zip file
• Comparison between working modes in the algorithm. zip file

Explanation of the contents

FAO test cases

The contents of the zip file are a set of test cases, each one separated in a folder. The test cases were used for populating the fisheries ontologies. These ontologies were developed for use within the Fish Stock Depletion Assessment System (FSDAS). All ontologies are publicly available from the FAO website. For each test case the following contents can be found:

• The ontology file, necessary for the inference.
• The mappings file, specified in R2O.
• The mappings file, expressed in a more schematic way, for the readers not familiarized with R2O.
• A comparison file, which contains the differences between the ontology file and the mappings file. The paper specifies the number of concepts, object properties and datatype properties in the ontology, the number of those that are covered and also the coverage as a percentage. The information presented here is more raw and detailed. In these files the elements that are present in only one of the two files are enumerated, which allows to check the specific elements that cause the partial ontology coverage from the mappings and possibly also vice versa.

Working modes comparison

The contents of the zip file are grouped by the mappings file used to test the algorithm, that is the meaning of the four folders, atlas, bcn and egm for the mappings files Atlas, BCN200 and EGM that map with Hydrontology, the folder phenom for Phenomenontology and the only mappings file considered for the evaluation with this ontology in the paper. In each one of these folders the contents are equivalent:

• An ontology file, Hydrontology for three of the atlas, bcn and egm. Phenomenontology for phenom.
• A mappings file depending on the folder, either Atlas, BCN200, EGM or the mappings for Phenomenontology.
• A plain text file with the query that is posed to the system, always named "queries.txt", the query is "Q(?0) <- Aguas(?0)" for the examples with Hydrontology and "Q(?0) <- Red(?0)" for the example with Phenomenontology.
• A set of plain text files containing the results of the execution. These files correspond to two different patterns:
  • resXY.txt where X is the working mode corresponding to the original REQUIEM system and Y is the working mode corresponding to the pruning methods developed in the context of the current paper. These files contain the execution trace, with all the clauses generated in each phase of the algorithm. Due to the verbosity to generate this trace the times may vary slightly when compared to the times specified in the paper.
  • resRQMXY-0 where X is the working mode corresponding to the original REQUIEM system and Y is the working mode corresponding to the pruning methods developed in the context of the current paper. The contents of the file show the final rewriting obtained as well as some additional information with respect to the test case.