This directory contains some datasets for VLog exercising the chase procedure, and, on installation, installs some more. Installing this package allows one to run chases on either VLog or RDFox, on either MacOS or Linux.
Installation is quite simple:
sh setup
should do the trick. This will get the chasebench repository and unzip its data files. It will also get and install VLog. Now you are ready to run either VLog or RDFox on these data sets.
There are scripts run-vlog and run-rdfox. Names are self-explanatory. These scripts take two parameters: the first one describes the chase type, either restricted or skolem, the second one contains a dataset name. For each dataset, the directory params-datasets contains a file, whose name represents the dataset. For instance,
sh run-vlog skolem LUBM-001
will run vlog, skolem chase, on the LUBM-001 dataset. Output will appear on a file vlog-LUBM-001.skolem.out. In addition, a summary will be printed on standard output: program used (VLog or RDFox), the chase type, the dataset, the date, what kind of processor, the runtime of the chase itself, the memory used, and the number of derivations.
During setup, the chasebench repository is downloaded. It is available from https://github.com/dbunibas/chasebench.git. It is described in: Michael Benedikt, George Konstantinidis, Giansalvatore Mecca, Boris Motik, Paolo Papotti, Donatello Santoro, and Efthymia Tsamoura. 2017. Benchmarking the Chase. In Proceedings of the 36th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems (PODS '17). ACM, New York, NY, USA, 37-52. DOI: https://doi.org/10.1145/3034786.3034796
VLog is available from https://github.com/karmaresearch/vlog.git.
The original database for Uniprot is available from http://www.uniprot.org. Reactome is available from http://reactome.org.