DPCat: Specification for an Interoperable and Machine-Readable Data Processing Catalogue Based on GDPR
Round 1
Reviewer 1 Report
The paper presents Data Processing Catalogue (DPCat) specification for representing and exchanging ROPA related information and provenance. In this paper, the Authors demonstrated the representation, querying, validation, and exchange of ROPA related information using DPCat and semantic web technologies. The Authors carried out the practical experiments using DPCat and verified it with SHACL to ensure the correctness of information based on legal and contextual requirements, and produce reports and ROPA documents based on DPA templates using SPARQL. The topic is interesting and the paper is well corresponding to the journal aim and scope.
The paper is well structured. The Authors formulated research questions and research objectives. The authors referred to the each of the research objectives in the conclusion section. Overall, the article was well prepared. As for the substantive part, I have no comments. The authors successfully established the extent to which the DPCat specification for an interoperable and machine-readable data processing directory based on DCAT-AP and DPV could overcome the heterogeneity of sources to facilitate the preparation of ROPA.
Minor typos:
Figure 8 – the contour line can be deleted.
Self-citations appear in the bibliography (10/38). That is quite a lot.
There are many abbreviations in the article, and more specifically in the abstract itself. It does not make the initial reading of the article easier.
Author Response
Reviewer #1
We thank the reviewer for their helpful comments. Below are our responses to specific statements.
R1: Figure 8 – the contour line can be deleted.
Response: (we assume by contour the border around the figure) We kept this for consistency in style with other figures which have that line.
R1: Self-citations appear in the bibliography (10/38). That is quite a lot.
Response: We have added additional relevant references to improve the ratio to 10/45. We also have mentioned the lack of directly relevant work to our research, which skewed the ratio in the earlier draft. The current list of self-citations is justified as follows:
Prior work that we build upon in this article: [2], [3], [5], [15]
Relevant work whose ouctomes we utilised: [6], [15], [40], [42]
Mentioned as relevant work in State of the Art section: [27], [29], [30]
R1: There are many abbreviations in the article, and more specifically in the abstract itself. It does not make the initial reading of the article easier.
Response: We have expanded the abbreviations on their first occurence in the abstract and the article.
In addition to the above, we have changed certain descriptions related to DCAT-AP for accuracy of how our work relates to it. For example, we clarify the relation between the two through "based on" rather than "extends as an profile" as we only intend to maintain compatibility with DCAT-AP rather than extend it.
Reviewer 2 Report
The paper is well organised in the domain of semantic web creation. However, it is intended for more clarification, consider a few tips below.
- Can you provide a list of a couple of competency questions for the proposed semantic model?
- The SHACL graph is already used in the work, but I think it is preferable to represent some data graphs that are related to the SHACL constraint.
- Can you provide the link for the generated ontology?
Author Response
Reviewer #2
We thank the reviewer for their helpful comments. Below are our responses to specific statements.
R2: Can you provide a list of a couple of competency questions for the proposed semantic model?
Response: We have provided an example of a competency question in Section 3.2. The rest of the competency questions were explored in prior work we have cited ([2], [3], [5], [15]), and can be obtained also from Appendix A where we provide a table of fields where the term is asked as a question, e.g. "What technical and organisational measures are applicable or used" was shortened to "Tech/Org measures" as we could not fit the entire question text within the limited available space on the page.
R2: The SHACL graph is already used in the work, but I think it is preferable to represent some data graphs that are related to the SHACL constraint.
Response: We have added a summarised example of the data graph (Listing 3).
R3: Can you provide the link for the generated ontology?
The specification documentation is available online at https://w3id.org/dpcat and the RDF serialisation and other resources are available in its repo at https://w3id.org/dpcat/repo
In addition to the above, we have changed certain descriptions related to DCAT-AP for accuracy of how our work relates to it. For example, we clarify the relation between the two through "based on" rather than "extends as an profile" as we only intend to maintain compatibility with DCAT-AP rather than extend it.
