Using Ontologies to Create Machine-Actionable Datasets: Two Case Studies

Round 1

Reviewer 1 Report

In paragraph 3 on page 2 the reference [5] should be moved left, immediately following the word 'FAIRification'.

The term 'units of measure' should be replace with 'units of measurement' throughout the document.

Please add and discuss the following important references to this topic area.
- https://doi.org/10.1186/s13321-016-0168-9
- https://i-adopt.github.io/

In Table 2, the QUDT entity referenced for the meter should be listed as 'M, applicableUnit instance of Length class' (see the QuantityKind vocabulary).

Finally, Figures 4 and 5 need to be redone to make things clearer:
- change the circles to boxes, to allow the text to inside
- change the unit in figure 4 to '1'
- unit is missing in the legend for Figure 4
- align the colors of the boxes between Figures 4 and 5.

Author Response

Dear reviewer,

we have revised our manuscript according to your comments.

1. We have improved the introduction with more scientific references related to SOSA and SSN, an additional paragraph on the origin of ontologies for units including 5 additional references, a reference to the latest (to the best of our knowledge) review of ontologies of units of measurement (Keil et al. 2018).
2. We have moved the requested reference.
3. We have replaced 'units of measure' with 'units of measurement', except when referring to UNECE, UCUM and OM 2.0, as the original materials use the term "measure" instead of "measurement".
4. Thanks for pointing out the two references, we have added them to the list of ontologies we want to use in our future work (second paragraph of the conclusion)!
5. We have added mentions to which quantity kind each QUDT unit is applicable, for 'M' in Table 2, but also all QUDT units in Table 1 for consistency.
6. We have remade Figures 5 and 6 (4 and 5 in the previous version of the manuscript).

Reviewer 2 Report

Dear authors,

I hope you will continue the work in this field and will publish further studies as you promise in the manuscript. The idea of the work is not new, but obviously you have achieved the goal you point out in the abstract.

I would like to emphasize that the very specific topic is discussed in the manuscript. The problems under discussion are rather philosophical than scientific. From this point of view, I think that is necessary to substantially enlarge the Introduction and to make clearer Materials and Methods. In this case, the manuscript will be more attractive for readers. In present form, the Introduction includes only 9 references, 5 of which are the websites. Among these 9 references only two can be considered as scientific publications (nos. 6 and 9).

Probably, it makes sense to describe the idea of the work in details to make the manuscript understandable for the wide range of readers.

Author Response

Dear reviewer,

we have revised our manuscript according to your comments.

1. 1. We have improved the introduction with more scientific references related to SOSA and SSN, an additional paragraph on the origin of ontologies for units including 5 additional references, a reference to the latest (to the best of our knowledge) review of ontologies of units of measurement (Keil et al. 2018). We have also added a scientific reference for the PROV-O ontology, however we couldn't find any peer-reviewed publication for QUDT, even early applied papers around 2011 cite the qudt.org web site.
   2. we have added a whole new subsection in Materials and Methods (2.5) that gives more details about the implementation workflow and the software tools we have used. We have also moved the paragraph on SPARQL formerly in the Discussion section to that new subsection for better readibility.
   3. We have made the source code and other supplementary materials available to readers on Zenodo here. 

Reviewer 3 Report

The paper shows two examples of using sematic web technologies to create machine - actionable datasets. The authors describes the FAIR concept for the data and create dataset which can be machine processed via queries. 

It is clear that authors created the tool and check the usability of it. For the reader is rather difficult to use the information, because better description of the system is missing. The supplementary files make it better, more information is needed.

Definitely, the need of human interface is important to the community and it will increase impact of the paper.

Author Response

Dear reviewer,

we have revised our manuscript according to your comments.

1. 1. We have added a whole new subsection in Materials and Methods (2.5) that gives more details about the implementation workflow and the software tools we have used. We have also moved the paragraph on SPARQL formerly in the Discussion section to that new subsection for better readibility.
   2. We have made the source code and other extra materials available to readers on Zenodo here.