OntoAndalus: an ontology of Islamic artefacts for terminological purposes

Tracking #: 2355-3568

Authors: 
Bruno Almeida
Rute Costa

Responsible editor: 
Special Issue Cultural Heritage 2019

Submission type: 
Ontology Description
Abstract: 
OntoAndalus aims at constituting a shared conceptualisation of the domain within a future multilingual terminological resource targeted at experts and students of Islamic archaeology. The present version of OntoAndalus is aligned with DOLCE+DnS Ultralite (DUL), an established top-level ontology for the Semantic Web. This article describes the modelling assumptions underlying OntoAndalus, as well as the more relevant design patterns (i.e. artefact types, events and individuals). The latter are exemplified through relevant case studies in the domain, namely those of lighting artefacts, the life cycle of pottery and the several descriptions of Vaso de Tavira.
Full PDF Version: 
Tags: 
Reviewed

Decision/Status: 
Accept

Solicited Reviews:
Click to Expand/Collapse
Review #1
Anonymous submitted on 08/Dec/2019
Suggestion:
Accept
Review Comment:

This paper is clear in its goal, formulation and explication. Applying the DOLCE+DNS Ulatralite (DUL) ontology and specializing it, the article outlines the work undertaken to create OntoAndalus, a shared conceptualization for al Andalusian pottery artefacts as a subdomain of Islamic archaeology. The approach undertaken is appropriate and follows well known and established methodology. The modelling activity departs from appropriate domain resources in order to ground itself empirically (textbooks, literature etc.) and applies the OntoClean method in order to critically judge the classes and relations proposed. The application of Dolce classes and reasoning is well articulated and logically consistent with that ontology. The article's application of DOLCE to the formulation of ontological descriptions of archaeological find types is clearly articulated and convincing. The explanatory quality of the text is high with the overall organization of the paper topics clear, and the illustration of the ontology done with legible and well explained diagrams. There are no obvious errors or unfounded claims in the work which references the relevant literature. For these reasons, I have no hesitation to recommend it for acceptance for publication.

Some comments for the authors follow.

The discussion around the class of ArchaeologicalArtefact correctly argues that this cannot be considered as a rigid class, since "No object is necessarily an archaeological artefact, only becoming one under certain conditions and for a certain period of its existence." This argument is also made in the articulation of CRMSci in the declaration of the 'Encounter Event' class and documented in related papers.

The discussion of the rigidness of artefact is interesting and the conclusion in line with positions taken in DOLCE and CIDOC CRM. That being said, the example used of the collected natural object which then becomes an 'artefact' shows the difficult boundary edge case which makes the distinction between 'artefacts' and 'natural objects' hard to sustain more broadly. This paper obviously need not resolve this ontological puzzle, but it surely deserves more thought.

The call for papers of this issue specifically references CIDOC CRM. The adoption and extension of DOLCE in this work, however, is excellent and clearly fits the case at hand. In particular the authors are clear to indicate that they look first to aid in the definition of understanding at a categorical level, and secondarily also to enable description of instances. It could be interesting in another work to discuss how it would be possible to gather data formulated in instance focussed ontologies (e.g. CIDOC CRM) and map/transform this data to put into relation with the categorical descriptions made possible through OntoAndalus.

Finally, one mention at the beginning of the paper is as follows: " It is thought that several observations made in this paper will lead to future research in the domain (e.g. artefact functionality, composition, uncertain knowledge in artefact descriptions)." While the attentive reader sees some of these emerge in the reading of the text, if this is an important point for the author(s) to make, they may wish to revisit this claim in the conclusion in order to point to what research may follow with references back into the text.

Review #2
By Michalis Sfakakis submitted on 29/Dec/2019
Suggestion:
Accept
Review Comment:

This manuscript was submitted as 'Ontology Description' and should be reviewed along the following dimensions: (1) Quality and relevance of the described ontology (convincing evidence must be provided). (2) Illustration, clarity and readability of the describing paper, which shall convey to the reader the key aspects of the described ontology.
------------------------------------------

This work reports on the development of the OntoAndalus ontology of Andalusian pottery artefacts, aiming to be used as a shared conceptualization by the domain experts and students of Islamic archeology. Ontology development is a challenging topic with many different aspects to consider. It requires the cooperation of different domain experts and even though it received plenty of attention, each specific development has its own challenges to deal with. The OntoAndalus Modeling assumptions are presented and the main design patterns of the ontology are exemplified. Furthermore, a short evaluation on technical aspects of the ontology deals with the consistency of the class and property hierarchies.

The development of the ontology is based on the interpretation of a corpus from Portuguese and Spanish domain specific texts, English textbooks and reference works, as well as from more specialized documents from related conferences and journals articles. The OntoAndalus version presented in this work uses DUL (DOLCE+DnS Ultralite) as the foundational ontology and OWL as modeling language.

The paper is well written and presents a thoroughly constructed approach. The authors clearly present their decisions for the development of the ontology, including core aspects and trends of the underlying domain of interest. The cases used to exemplify the main representation patterns are very informative for both modeling decisions and the use of OntoAndalus.

While the authors reported shortly for the technical evaluation of the ontology, the paper should be strengthened with more details on the evaluation results from use of the HermitT reasoner in Protégé and the application of the OntoClean method.

Review #3
By Kalliopi Kontiza submitted on 28/Apr/2020
Suggestion:
Accept
Review Comment:

This manuscript was submitted as 'Ontology Description' and should be reviewed along the following dimensions: (1) Quality and relevance of the described ontology (convincing evidence must be provided). (2) Illustration, clarity and readability of the describing paper, which shall convey to the reader the key aspects of the described ontology.

The paper "OntoAndalus: an ontology of Islamic artefacts for terminological purposes" presents the aligned with DOLCE+DnS Ultralite (DUL) OntoAndalus ontology, part of a future multilingual terminological resource targeted at experts and students of Islamic archaeology. The article describes the modelling assumptions underlying the proposed ontology, as well as the more relevant, interesting and challenging design patterns (i.e. artefact types, events and individuals) of such attempt. The case study of Vaso de Tavira attempts to exemplify these design patterns, namely those of lighting artefacts, the life cycle of pottery and the several descriptions of the artefact.

The aim of the paper was to showcase how the conceptualisation of the domain of pottery artefacts (Andalusian) and the future onto-terminological resource can help the domain experts in the field of Islamic Archaeology to meet the challenges with regard to terminological harmonisation and advance further interdisciplinary endeavours in ontology creation in the field of archaeology.

The critical and informative sections of 2. Modelling assumptions, where the paper expands on the modelling aspects of ‘Classes and collections’, ‘Parts and dependent places’, ‘Qualities and attributes’, and 3. Main design patterns, where details for one of the Artefact types the lighting artefact class are presented, highlight the challenge of creating comprehensive categories due to the diversity of descriptions provided by the experts. Section 2 provides comparative information between DOLCE, CIDOC and the proposed ontology with an interesting focus on the relationship between the members or parts of an object. Especially the structure of section 3 allows the reader to follow the design process and how the above mentioned challenges were tackled along the way; it highlights for the reader the contribution of the paper to the interested research community, since discussion around the concept of artefact applies to similar challenges faced by other researchers and subfields or applied research of the broader Cultural Heritage domain.
As a minor comment, it would be interesting to have more details or thoughts (Conclusion section) on how the developed ontology would be integrated within the discussed future onto-terminological resource.

Other Minor issues

Reference [20]: a link pointing to the CIDOC-CRM version mentioned would be helpful if referencing style allows it. Same for references [31,32].

Figure 6: it would make the Figure more self explanatory if the terms of the objects were added at the caption line (i - iv)

Figures 12 and 13: there is a connection between these two pictorial representations which is not clear to the reader. Maybe a colour coding scheme or a reference from Figure 12 to Figure 13 could facilitate this.