This project is used for merging multiple results from fagi instances after using the partitioning process.
Clone the project to a preferred location:
git clone https://github.com/SLIPO-EU/fagi-merger.git fagi-merger
Then, go the root directory of the project (fagi-merger) and run:
mvn clean install
Go to the target directory of the project and run:
java -jar fagi-merger-SNAPSHOT.jar -config /path/to/config.xml
Inside the resources directory of the project there is a config.template.xml file.
partitions
the number of partitions produced during the partitioning process.
fusionMode
the fusion mode used for the fagi instances.
datasetA
the path of the first dataset.
datasetB
the path of the second dataset.
unlinkedA
the path of the file that the non-linked entities from dataset A will be written. The file gets created if it does not exist (unlike outputDir).
unlinkedB
the path of the file that the non-linked entities from dataset B will be written. The file gets created if it does not exist (unlike outputDir).
inputDir
the directory path that contains the partitions.
partialOutputDirName
the directory name with the fusion output under a partition directory.
target
This tag contains child tags (outputDir, fused, remaining, ambiguous, statistics, fusionLog, fusionProperties) with the output configuration of the merging process.
outputDir
the output directory path for the merged results.
fused
the output filepath of the (merged) fused file.
remaining
the output filepath of the (merged) remaining file.
ambiguous
the output filepath of the (merged) ambiguous file.
statistics
the output filepath of the computed statistics file.
fusionLog
the output filepath of the (merged) fusion log file.
fusionProperties
the output filepath of the (merged) fusion properties file.