Stones Heritage Ontology

Against the background of the INT-ACT project, the paper develops the Stones Heritage Ontology (SH Ontology), implemented in OWL, as an extension of the CIDOC Conceptual Reference Model (CIDOC-CRM) with the aim of representing both tangible and intangible entities in the cultural heritage domain. The SH ontology is illustrated with two examples: Cromeleque das Fontainhas in Portugal and Calanais I in Scotland.

On the one hand, I find it worthwhile to develop an ontology of stones heritage. In addition, the data file appears to be legitimate according to the general criteria of assessment, although this task might go somewhat beyond the scope of my expertise. On the other hand, I should say that the paper in its current version will require substantive revision to be accepted as a full paper in the Semantic Web journal (SWJ). (Remark: “[Full papers] will be reviewed along the usual dimensions for research contributions which include originality, significance of the results, and quality of writing”, extracted from: https://www.semantic-web-journal.net/reviewers .)

First of all, the originality/novelty of this work may be unclear, in particular as compared to previous studies about CIDOC-CRM. For one thing, there are many existing works on the mapping, merging and extending of CIDOC-CRM (see e.g., [1] https://doi.org/10.3390/heritage2010040 for a survey). Moreover, the potential benefits of applying machine learning techniques to CIDOC-CRM have been explored (see e.g., [2] https://doi.org/10.3390/heritage5030084 for a survey). Although the authors discuss several related works on this ontology in Section 2, a comprehensive and critical examination of them may not be provided and therefore the development of the SH Ontology may not be sufficiently motivated or justified.

For another, even though CIDOC-CRM may play a crucial role in the cultural heritage domain (see [1] for details), it has been argued that the foundations of CIDOC-CRM require strengthening from logical and ontological points of view (see e.g., [3] https://doi.org/10.1504/IJMSO.2018.098393 and [4] https://doi.org/10.3233/FAIA200664 ). More specifically, the application of CIDOC-CRM to the archeological domain may warrant close scrutiny of the existing classes of this ontology. Consider for instance the proposed class Archaeological Artefact (“Artefact found in an archaeological site”) which is a subclass of E24 Physical Human Made Thing (page 11). Consequently, the SH Ontology implies that any archaeological artefact is “purposely created by human activity”. However, according to the SWJ paper developing an ontology of Islamic artefacts ([5] https://doi.org/10.3233/SW-200387 ), an archaeological artefact is something that is, for example, “recognised as an evidence of material culture in an archaeological site” [5, p. 297] and archaeological artefacts should be distinguished from “designed artefacts”. While remaining agnostic as to the validity of this conceptual analysis of archaeological and designed artefacts, I find the paper lacking in such a meticulous conceptual analysis of tangible and intangible cultural heritage, and hence the significance of the proposed SH Ontology may not be well supported.

Overall, the paper has the potential to contribute to the ontology of cultural heritage; however, it requires major revisions to be suitable as a full paper for the SWJ.

[Minor points]
- Running title: “Stones Ontology” -> “Stones Heritage Ontology”
- Terminological consistency: “model” versus “ontology”. For instance, as far as CIDOC Conceptual Reference Model (CIDOC-CRM) is treated as an ontology for cultural heritage, it would avoid confusion to use the term “ontology” to describe CIDOC-CRM throughout the paper.
- Notational consistency: “INT-ACT” versus “INTACT” (e.g. page 2) and “CIODC-CRM” and “CIDOC CRM”.
- Notational clarity: It would enhance readability to italicize class names such as “E2 Temporal Entity”.
- Page 1, the first sentence of the text: It is good to add a reference about the INT-ACT project.
- Page 1, line 46: “propose” -> “proposes”.
- Page 1, the last sentence: Since this is the first time to mention CIDOC-CRM in the text, it would be better to say “CIDOC Conceptual Reference Model (CIDOC-CRM)” with some appropriate reference(s).
- Page 2, line 34: The text “The new concepts … with the two studied sites” may require restructuring to be more readable.
- Page 2, line 36: “presents final remarks and future works”. Better to say, for example: “provides a conclusion along with some remarks on future work”.
- Page 25, line 47: “6. Final remarks” -> e.g., “6. Concluding remarks”.
- Section 6: I do not think that the authors discuss clearly future directions of research.

The paper is fairly well written, although it would benefit from some editing, particularly with regard to paragraph structure and the presentation of tables. The author presents an ontology, the Stone Heritage (SH) ontology, implemented in OWL, with applications in the cultural heritage domain, specifically within archaeology. The contribution is particularly relevant for managing data in connection with technologies related to extended reality experiences. Overall, this is a welcome submission for the journal, and I believe it could be accepted after major revisions. Below, some remarks and suggestions are provided.

1) Originality:

Although the scope and aims of the paper are clear, the originality of the work with respect to the state of the art is not entirely evident to me. Let me clarify my concerns.

At pp. 3-4 the authors make two major claims:

i) the SH ontology is helpful for modeling both tangible and intangible entities in the cultural heritage domain;

ii) the SH ontology is an extension of the CIDOC-CRM ontology.

Regarding (i), at p. 8 the authors further claim that the SH ontology is built along two main semantic dimensions, namely the tangible and intangible. Now, the notion of the intangible cannot be fully explained simply as the opposite of physical or material. This difficulty makes it challenging to formally capture such a category, which remains somewhat vague and debated in the cultural heritage community (as one can infer from Paragraph 2 on related work). Yet, the authors do not deal directly with this distinction in the ontology itself, for instance by introducing formal constraints. Instead, they state that “these dimensions are not part of the ontology itself; they are didactic backbones for both the development and the presentation of the ideas behind the ontology” (p. 8).

I suggest moving this statement about the didactic use of these dimensions to the introduction, and clarifying more explicitly in what sense the SH ontology employs, albeit informally, the tangible/intangible distinction compared to the state of the art. For example, the authors mention Qingyang’s study, in which images, videos, or audio files are linked into the ontology rationale to represent intangible aspects. What is the authors’ position in this respect? To me, this does not amount to a genuine ontological characterization of the intangible, which as a category is somewhat abstract. Rather, these multimedia elements appear to provide additional context for referring to the notion of ‘intangible’.

Regarding (ii), the claim that the SH ontology is an extension of CIDOC-CRM needs to be substantiated by specifying which set of axioms has been introduced, and why they qualify as an extension. Without such a formal specification, it is difficult to assess in what sense CIDOC has been extended. It may be that the authors intend “extension” in a looser sense, namely that the main classes of CIDOC are specialized via subclass relations to address the archaeological domain.

If this is the intended meaning, then I would consider this point as the main contribution of the paper: an application of CIDOC in the archaeological domain. In either case, the paper would benefit from providing further details on how the SH ontology addresses the open issues surrounding the tangible/intangible dimensions, and from making a clearer comparison with the state of the art and also the ontologies related to immersive extended reality.

2) Significance of the results
Overall, the backbone of the ontology is well structured and effectively illustrated through the two case studies, Cromeleque das Fontainhas (Portugal) and Calanais I (Scotland). These studies allow the reader to grasp the main ontological engineering choices behind SH and its potential for managing the data it was designed to handle.
With regard to the case studies, however, the historical introductions in Paragraph 4 strike me as somewhat too long for an ontological paper. I suggest shortening these sections in favor of a more concise presentation of the main classes and relationships.
The paper also includes a large number of tables (25 in total), each providing brief descriptions of entities. While detailed, the number of tables risks overwhelming the reader and obscuring the overall rationale. Since readers interested in full details can directly inspect the ontology in an editor such as Protégé, there is little need to report the entire ontology in the paper. It would be more effective to present only the most relevant classes and relations, highlighting how they address practical problems in the domain.
Additionally, the ontology could benefit from some simple yet effective logical constraints. For example, the classes BronzeAge and Neolithic are not explicitly declared as disjoint. This means that if an entity currently classified as Neolithic is later found (as a consequence of further studies, as often happens in history) to belong to BronzeAge, and the ontology is not updated accordingly, the entity will simultaneously fall under both classes without triggering any inconsistency for the reasoner.

3) Quality of writing:
The paper is generally well written and logically organized. However, as mentioned earlier, I suggest reducing the number of tables and also figures (currently 13 figures). I would also move and integrate paragraph 6.1, which is currently in the conclusions, into the introduction or related work.
In the conclusions, some statements would benefit from further clarification rather than being simply stated.
For example, the sentence “These layered elements offer a richer digital representation aligned with concepts of Digital Twins” could be elaborated to explain such alignment with digital twins. Similarly, the statement “The environment is the first E of the 3E approach, which integrates environment, experience, and emotion. Tangible aspects are therefore more detailed” could be analyzed in greater depth. The 3E approach seems to be an interesting notion, yet it is only presented in the introduction and briefly summarized in the conclusions. Expanding on it would provide more insight for the reader.

4) Long-term stable URL for resources
The authors indicate that they are working on providing a stable URL to make the ontology fully accessible upon acceptance of the paper.
I would suggest adding some annotations to the ontology to clarify the engineering rationale.

This manuscript presents an ontology for Stone Monument Heritage, modeled as an extension of the CIDOC CRM family. The paper is generally clear, well structured, and supported by illustrative figures and tables. The topic is relevant for the semantic modeling of cultural heritage and aligns well with ongoing efforts in ontology-based documentation of archaeological and monumental heritage. However, several conceptual, methodological, and presentation issues limit the strength of the contribution in its current form.

(1) Originality

The proposal of an ontology specifically focused on Stone Monument Heritage represents a potentially original contribution, particularly insofar as it attempts to formalize monument-specific concepts within the CIDOC CRM ecosystem. However, the manuscript does not sufficiently position itself with respect to existing CIDOC CRM extensions, especially CIDOC CRM Archaeo, nor does it clearly articulate what modeling gaps remain unaddressed by existing standards.

In particular, the related work section does not adequately justify the decision to develop a new extension rather than reuse or adapt CIDOC CRM Archaeo. Without a clearer comparison and rationale, the originality of the contribution remains somewhat weakened, as it is unclear whether the proposed ontology introduces genuinely new conceptual insights or mainly reimplements existing CRM constructs.

(2) Significance of the Results

The potential significance of the results lies in providing a structured semantic framework for describing stone monuments and associated activities. Such a framework could be useful for data integration, interoperability, and long-term documentation in heritage contexts.

However, the impact is reduced by several modeling inconsistencies and design choices that are not sufficiently justified. For example:

Several newly introduced classes (e.g., excavation, identification) appear to overlap semantically with E7 Activity in CIDOC CRM and could be modeled using existing constructs.

The introduction of a Time-Span class raises concerns, as it appears to duplicate E52 Time-Span without a clear explanation of the added value.

Roles such as archaeologist or volunteer are modeled as classes, whereas they could more appropriately be represented as roles within the CIDOC CRM logical framework.

The monument Cromeleque das Fontainhas is not modeled as an E53 Place, which prevents the use of standard topological relations. Given existing work on spatial and topological modeling in CIDOC CRM (e.g., RCC8-based approach), this modeling choice requires justification or revision.

Without resolving these issues, the contribution risks being perceived as conceptually redundant or misaligned with established CIDOC CRM modeling practices.