Benchmarking system extension (datasets)

[dersuchendee]

The benchmarking system can be extended with two dataset sources:

1. Data derived from related work
2. Data derived by the WHOW and ArCO European projects (EU directives, ontologies)

[Cremadvd]

Upload of new CQs dataset
The attached file is structured as the original competency question dataset and contains 235 additional CQs from five different projects: SWO, STUFF, DemCare, AWO and Ontodt.
Let us know if it is enough to complete the task or if we need to update the original dataset.

full_cqs.csv

[dersuchendee]

Thank you! It would be helpful to know whether you used specific related work to compile this, e.g. Alharbi et al. and Rebboud I remember them using some of these datasets.

For the KGs we used as reference, it would be great to add to the original dataset (or also put them here and we can integrate) links to related ontologies (water monitoring and health monitoring, findable here for WHOW (link to the OWLs directly, raws, uploaded online) and an additional column called "support documents" where if there are any EU regulations it is possible to add them (e.g. for WHOW it is "Directive 2000/60/EC establishing a framework for Community action in the field of water policy", findable online).

We could also add for instance links to this material which corresponds to the following ground truth CQs: https://github.com/XueliPan/GenCQs/blob/main/hci/ground-truth-cqs.txt.

The csv is enough, you can paste it here and I can integrate it, thanks!

[Cremadvd]

Source specification
The paper detailing the dataset can be found at the following link: https://www.sciencedirect.com/science/article/pii/S2352340919314544

Inside the paper, there is a link to the entire dataset:
https://data.mendeley.com/datasets/pp6hmfxgfg/1
The submitted dataset was built by merging the information inside the Data folder, containing dataset descriptions, CQs, and SPARQL queries. The related ontologies can be found inside the Ontologies folder.