global-identifiers

core.sql contains all the relevant tables and stored procedures to setup the Virtuoso Conductor. Simply execute the contents of the SQL file.

The Maven project DBpediaIndentifierClustering contains a java project for clustering and rewriting. See the comments in the main method of InMemoryClustering.java for further details.

Data file rewriting

Run the rewriting with

mvn exec:java -Dexec.mainClass="DataRewriter" -Dexec.args=""

Possible Dexec.args:

-con followed by the connection string to the database.

-in followed by the path to the folder containing the input data. Default is the data path

-out followed by the path to the folder for the data output. Default is the data/out path

-limit followed by the number of lines to rewrite per input file. Default is no line limit

-clustering followed by the name of the clustering to use in the rewriting process.

-buffer followed by the desired size of the local identifier cache. Default is no limit.

-namespace followed by a commalist of namespaces. Renaming will be restricted to the specified namespaces.