Conceptual Modelling of Subjectivity, Temporality and Vagueness with ConML

Abstract
Research and practice in the humanities often involve the management of large and complex bodies of information representing objects of study, hypotheses, work in progress or even conclusions. Although natural language is usually employed to convey these, semi-formal languages have shown to be useful in a number of situations due to reduced ambiguity and ease of implementation on computing platforms (CIDOC, 2011; Gonzalez-Perez, 2012). Conceptual modelling, the technique to construct semi-formal models of the world that can be understood by humans and machines, is thus an excellent approach to fight the complexity that very often makes humanities a difficult endeavour. By using conceptual modelling, we can explore unknown areas of research, document entities in the world for latter reference (in the form of inventories or catalogues, for example), communicate complex situations and processes to others (in our own or a different discipline), prescribe how computer systems should be constructed (very much like an architectural drawing prescribes how a building is to be built), and facilitate the interoperation of heterogeneous bodies of data through the gradual refinement of models (Gonzalez-Perez and Martín-Rodilla, 2015).
Within conceptual modelling, however, some areas are still poorly understood and weakly supported by existing technologies. One of these areas is that of “soft issues”, that is, the modelling of issues that are crucial to human understanding, but which have been traditionally regarded as incompatible with or almost intractable by computers. Some examples include the passage of time, the subjective perception of the world, and vagueness.
As these issues are usually difficult to implement as part of computer systems, they are often left out of formalised representations, which results in models that become too simplistic to be useful. For example, very few databases in the humanities are capable of recording information diachronically (i.e. representing changes over time), subjectively (i.e. recording the voices of different relevant agents) and vaguely (i.e. acknowledging that the world is not clear-cut, and that fuzzy boundaries and imperfect knowledge exist). However, soft issues are a crucial component of research and practice in digital humanities, so they should not be left out.
In this tutorial, we use ConML (www.conml.org) to demonstrate how these issues can be tackled, and how conceptual models can be constructed that express temporality, subjectivity and vagueness. ConML is a conceptual modelling language developed with a special orientation towards the humanities and social sciences, and with the aim that specialists with little or no experience in computer science would be able to learn it and use it in very little time (Gonzalez-Perez and Martín-Rodilla, 2017). ConML has been successfully used in a number of projects, such as the development of the Cultural Heritage Abstract Reference Model (CHARM, www.charminfo.org) (Incipit, 2016). Specifically, ConML supports the expression of temporal, subjective and vague aspects of information through a number of mechanisms, which will be presented and exercised in the tutorial.
The tutorial will employ theoretical explanations combined with hands-on group exercises, a technique that we have been since 2012 in our postgraduate courses on information modelling for the humanities at the University of Santiago de Compostela. Attendees will acquire practical skills to construct expressive models, as well as the theoretical underpinnings supporting them. They will have access to proven modelling techniques as well as new and experimental findings.
By incorporating the modelling of soft issues into their toolbox, attendees will be able to construct conceptual models that document the object of study much better, communicate intentions in a richer manner, and allow the implementation of computer systems capable of catering for these complex but essential aspects of the humanistic inquiry.

Outline
The tutorial will be organised as follows.
1. The need to represent. Benefits of conceptual modelling. 15 minutes.
2. Introduction to ConML. Types and instances. Classes, enumerated types, attributes, associations, generalization relationships. Basic modelling exercise, in groups. 1 hour.
3. Expressing subjectivity. Theory of disagreement. Subjective aspect classes. Existence and predication qualifiers. Modelling exercises. 30 minutes.
4. Expressing temporality. Theory of change. Temporal aspect classes. Existence and predication qualifiers. Modelling exercises. 30 minutes.
5. Expressing ontological and epistemic vagueness. Dealing with imprecision. Dealing with uncertainty. Using inaccuracy. Modelling exercises. 30 minutes.
6. Wrap up and conclusions. 15 minutes.

Target audience
The tutorial is targeted at researchers, instructors, students or practitioners of digital humanities, including anthropology, archaeology, art studies, history, law, linguistics, literature and other disciplines, who need to manage data or information as part of their job and feel an inclination to semi-formal approaches to knowledge representation and management. The tutorial does not assume previous knowledge of conceptual modelling, but some experience with information management systems and languages will help. Ideally, attendees should be between 10 and 15.

Instructors
Dr. Cesar Gonzalez-Perez is Staff Scientist with Incipit CSIC in Spain. He leads a co-research line in software engineering and cultural heritage. His main interest is to understand and assist the knowledge-generation processes that occur in relation to cultural heritage.
Dr. Patricia Martín-Rodilla is a Postdoctoral Researcher with CITIUS at the University of Santiago de Compostela. She works on knowledge extraction from textual sources and the representation of discursive knowledge in the humanities.
Both Cesar and Patricia have extensive experience in teaching conceptual modelling in the humanities, having collaborated in the design, teaching and management of postgraduate courses in this field since 2012, as well as over 10 related workshops at conferences over the last few years.

Bibliography
CIDOC (2011). The CIDOC Conceptual Reference Model (web site). Accessed on 26 November 2012. http://www.cidoc-crm.org/
Gonzalez-Perez, C. (2012) A Conceptual Modelling Language for the Humanities and Social Sciences. In Rolland, C., Castro, J. & Pastor, O. (Eds.) Sixth International Conference on Research Challenges in Information Science (RCIS), 2012. IEEE Computer Society. 396-401.
Gonzalez-Perez, C. & Martín-Rodilla, P. (2015) Integration of Archaeological Datasets through the Gradual Refinement of Models. In Giligny, F., Djindjian, F., Costa, L., Moscati, P. & Robert, S. (Eds.) 21st Century Archaeology: Concepts, Methods and Tools - Proceedings of the 42nd Annual Conference on Computer Applications and Quantitative Methods in Archaeology. Archaeopress. 193-204.
---- (2017) "Teaching Conceptual Modelling in Humanities and Social Sciences". Revista de Humanidades Digitales, 1, 408-416.
Incipit (2016). CHARM White Paper, version 1.0.6. Incipit, CSIC. http://www.charminfo.org/Resources/Technical.aspx. 30 May 2016.