link-based-complementary-fusion

A light-weight tool to fuse complementary facts to DBpedia identifiers

Usage:

Under the directory created from downloading/cloning, build the project and produce an executable jar using

$ mvn package

The resulting target directory contains jar file JRCNamesJena-0.0.2-SNAPSHOT.jar that include the project and dependency files.

Use the following format to run it

$ java -jar ./target/JRCNamesJena-0.0.2-SNAPSHOT.jar --query <path/to/query.rq> [--limit <page size>] [--outdir <output files directory>] [--outfileprefix <prefix>] --sparqlendpoint <URI to SPRAQL endpoint>

where the arguments are:

• query A SPARQL query of CONSTRUCT type, which is used to to match data against DBPedia.

• sparqlendpoint URI to the SPARQL endpoint

• limit Required page size; Defaults to 10000, actual page size should be checked against the SPARQL endpoint.

• outdir Directory where output (paged) files are written, defaults to out under the user home directory

• outputfileprefix Prefix for the output files. Each file will additionally have a number which corresponds to the offset of records from the SPARQL endpoint; defaults to p .

To link DBPedia resources of persons and organizations to their corresponding entries in JRC-Names use:

$ java -jar ./target/JRCNamesJena-0.0.2-SNAPSHOT.jar --query ./resources/jrcnames-fuse.rq --limit 10000 --outdir ./out --outfileprefix p --sparqlendpoint http://data.europa.eu/euodp/sparqlep

(Values of outdir and outfileprefix can be of course different than those suggested above)