FAQ

Q: What data formats can the SDM-RDFizer convert? A: The SDM-RDFizer can convert the following data formats:

• CSV
• JSON
• XML
• MySQL
• Postgres

Q: How are the transformation rules defined? A: The SDM-RDFizer follows the transformation rules defined by RML. (https://rml.io/specs/rml/)

Q: How is the SDM-RDFizer configured? A: The SDM-RDFizer requires a configuration file. The configuration file has the following fields:

• main_directory: The folder where the mapping and data sources are located.
• number_of_datasets: The number of datasets to be converted.
• output_folder: The location where the output will be generated.
• remove_duplicate: Remove duplicates from the output.
• all_in_one_file: If there are multiple datasets, the SDM-RDFizer can generate only one output file.
• name: Name of the output file.
• enrichment: When removing duplicates the SDM-RDFizer has two settings. If set to “yes”, the SDM-RDFizer uses hash tables for the storage of the generated triples (best performance). If set to “no”, the SDM-RDFizer uses arrays for the storage of the generated triples.
• mapping: The location of the mapping.

Example of a configuration file: https://github.com/SDM-TIB/SDM-RDFizer/wiki/Install&Run

Q:How can I run the SDM-RDFizer? A: We have two ways of running the SDM-RDFizer:

• From the terminal: python3 rdfizer/run_rdfizer.py /path/to/configfile
• Docker container:

1. Building the docker container. Note: All documents in the same folder of the Dockerfile will be copied to the container. docker build -t rdfizer .
2. To run the application, you need to map your data volume to “/data” folder of the container where data, mappings, and config files should be located: docker run -d -p 4000:4000 -v /path/to/yourdata:/data rdfizer
3. Send a POST request with the configuration file to RDFizer the file curl localhost:4000/graph_creation/data/your-config-file.ini
4. Get the results from the container (if output folder is inside the data folder, results are already in your host) docker cp CONTAINER_ID:/app/path/to/output .
5. Instead of installing from scratch, you can use the docker image we have created.

Q: Where can I take a look at the experimental evaluations ran over SDM-RDFizer? A: The experimental evaluation can be seen here: https://github.com/SDM-TIB/SDM-RDFizer-Experiments/