Dear Mika,
On 22/2/2016 2:01 μμ, Mika Nyman wrote:
Personally I have taken a special interest in questions related to
subjectivity. Brutus dagger meeting Caesar in a space-time volume is a
striking example. I think it is imperative to try to reach to the
intentions of Brutus through what is documented in literature. In my
view it is a duty in building digital environments for cultural
heritage to reach out to sources that could enlighten the intentions
of actors and place them in a broader context. (I mention intentions
just as one aspect of subjectivity.) I think it is a duty of the CIDOC
CRM SIG to point out to those developing strategies for new digital
environments that these kinds of links, including links to
subjectivity, can me modelled conceptually. What can be modelled
conceptually can also be implemented technologically.
I fully agree that what is really interesting in cultural research is of
course this, and not so much the physical part.
My current research interest is understanding this discourse, can it be
structured? Where are the limits
of knowledge representation methods, where are they helpful in this
discourse? How does a scholarly argument
work? Steffen Hennicke has analyzed in his PhD hundreds of natural
language questions of historians to archives,
and matched against an extension of the CIDOC CRM :
Title: "What is the Real Question? An Empirical-Ontological Approach to
the Interpretative Analysis of Archival Reference Questions"
Philosophische Fakultät I, Humboldt-Universität zu Berlin. I hope it is
published by now.
Indeed it appears that such sort of analysis can help to create better
documentation guide lines.
Currently, we work on the abstracts from EAA in Plzen to understand
archaeological questions empirically. We hope to
publish soon the results.
In both works, a layering of arguments starting from material facts to
mental phenomena, from individual facts to
collective behavior appears, as well as a relatively compact set of core
patterns of interest, which can be modelled
ontologically to a certain degree.
"What can be modelled conceptually can also be implemented
technologically" is a hard question. Recent research
in AI seem to show that human argumentation is dominated by relevance
estimations which are still not understood
in a formal way, and that even simple tasks become intractable without
such functions.
Neural networks are getting better, but they do not easily provide a
conceptual insight in what they have "learned".
Best,
Martin
--
--------------------------------------------------------------
Dr. Martin Doerr | Vox:+30(2810)391625 |
Research Director | Fax:+30(2810)391638 |
| Email: [email protected] |
|
Center for Cultural Informatics |
Information Systems Laboratory |
Institute of Computer Science |
Foundation for Research and Technology - Hellas (FORTH) |
|
N.Plastira 100, Vassilika Vouton, |
GR70013 Heraklion,Crete,Greece |
|
Web-site: http://www.ics.forth.gr/isl |
--------------------------------------------------------------
