=Paper=
{{Paper
|id=Vol-1117/paper5
|storemode=property
|title=Representation of Archival User Needs using CIDOC CRM
|pdfUrl=https://ceur-ws.org/Vol-1117/paper5.pdf
|volume=Vol-1117
|dblpUrl=https://dblp.org/rec/conf/ercimdl/Hennicke13
}}
==Representation of Archival User Needs using CIDOC CRM==
https://ceur-ws.org/Vol-1117/paper5.pdf
Representation of Archival User Needs using
CIDOC CRM
Steffen Hennicke1
Berlin School of Library and Information Science,
Humboldt-Universität zu Berlin, Germany
steffen.hennicke@ibi.hu-berlin.de
Abstract. This paper stems from an ongoing dissertation project and
demonstrates how the CIDOC CRM is used to create an ontological
model – the Archival Knowledge Model (AKM) – of common patterns
found in written natural language questions to archives. Such an onto-
logical model can be used to query archival or historical knowledge bases
in order to provide more adequate answers and to enable more relevant
discovery facilities. For this purpose, 330 reference questions to the Ger-
man Federal Archive are being analyzed and patterns found translated
to the CIDOC CRM and appropriate extensions. In particular, the pa-
per introduces the methodological approach to the interpretation of user
questions and the draft of a prominent pattern called Documentation-
Activity.
Keywords: CIDOC CRM, archival reference questions, access to archives,
archival user needs, Archival Knowledge Model
1 Introduction
The main means for discovering [1] and accessing primary sources in an archive
are finding aids and holding guides supported by the expertise of archivists.
These archival aids are descriptive tools which are meant to help the user to lo-
cate and discover relevant materials in the enormous and ever growing amounts
of rich information potentials [2] in archives. The conceptualization of these de-
scriptive tools as well as respective digital encoding standards like the Encoded
Archival Description 1 (EAD) are based on elaborated and historically grown
archival principles and models but their design is less informed by explicit knowl-
edge about the information needs of archival users [3] due to a prevailing lack
of qualitative in-depth analysis of archival user needs [4][5]. Such studies are,
however, a crucial cornerstone for the improvement of existing and future digital
archival information systems [6].
The hypothesis on which this paper rests is that it is not necessary, and even
not desirable, to change the archival description itself and related metadata
schemas, but, instead, it is possible, in principle, to supplement existing archival
1
http://www.loc.gov/ead/
and historical knowledge bases with an ontological model which matches typical
patterns from user inquiries to archives. Such an ontological model would make
knowledge explicit and add relevant context which is necessary to adequately an-
swer typical user questions and to create better discovery systems. Furthermore,
such ontological models enable empirically qualified assessments of metadata
schemas for archival information systems but also of archival cataloging rules.
The general research question, therefore, is if there is a hypothetical ontology
which can represent user inquiries and their probable interpretations as formal
queries against a model of the archival target world that would adequately an-
swer the inquiry or its implicit purpose. The result of this analysis is an onto-
logical model which represents inquiry patterns of different abstraction levels to
archives in the form of queries to this ontology. The CIDOC CRM2 has been
chosen as the ontological target model mainly for its strong empirical founda-
tion and event-based conceptualization of historical processes. Written reference
questions 3 from the German Federal Archives, the Bundesarchiv, have been cho-
sen as research data. This type of research data has been largely neglected in the
analysis of user needs in the archival domain, although they document a mostly
unfiltered information need in the users own words.
A brief literature review will establish the general research context followed
by a short introduction of the research data and the methodological approach
to the analysis of questions.4 The focus of this paper lies on the demonstration
of the interpretative translation of natural language questions to a common on-
tological representation. Two examples will demonstrate how shared patterns
in user questions and their probable interpretations can be translated to an
ontological model, the Archival Knowledge Model (AKM), covering and extend-
ing the CIDOC CRM. The specific pattern presented in this paper is called
Documentation-Activity which proposes two new classes as extensions to the
CIDOC CRM. It is important to bear in mind that all results presented in this
paper are preliminary and research is ongoing.
2 Research Context
The limitation on ”simple answers to clear cut, search term-based questions” [7]
is one of the core problems of today’s information systems on the Web. Pattern-
oriented retrieval could describe many more complex questions whose answers
go beyond the capacity of simple querying [8]. This limitation poses a significant
2
http://cidoc-crm.org/
3
The term reference question refers to a request of a user to a staff member of a
library or archive for information or assistance regarding the provision of any kind
of information. Such a request can either be posed in person at a reference desk or
remotely by phone, mail, or e-mail. In this study, only written reference questions
by mail or e-mail are being analyzed.
4
For more details on the dissertation, please confer the extended abstract which will
be presented at the Doctoral Consortium of the TPDL 2013 and published in the
conference proceedings.
barrier to more sophisticated and integrated information systems. Part of this
problem is a prevalent focus on traditional library cataloging and methodology
in describing and contextualizing objects of interest. At the same time, today
”the key challenge of organizing information is to construct systems that aid un-
derstanding, contextualizing, and orienting oneself within a mass of resources”
where models help to bridge a semantic gap between the formalizations in in-
formation systems and the conceptualizations of scholars [9]. Instead of a Global
Knowledge Network [7], mostly isolated ”silos of information” exist which all
employ their own idiosyncratic structures and data encodings. The Semantic
Web addresses these issues in its research agenda. However, this agenda suf-
fers from an ”almost exclusive focus on ’terminology’ rather than ’ontological
structures’” resulting in the neglect of fundamental and complementary lines
of research [7]. One such missing line of research is how typical user questions
are formally structured. The systematic and in-depth analysis of original user
questions from different stages of the research process is important and has the
potential to provide, for example, necessary information on query mechanisms
or adequate granularity of ontologies [7].
Discovery is one of the most important and re-occurring stages in research
processes especially distinctive for historical inquiry in archival settings. As al-
ready mentioned, research in the archival domain exhibits a lack of in-depth user
studies [10]. The study of Duff and Johnson [11] is one of the few which looked
at the type and structure of user reference questions to archives.5 Regarding
the domain of historical research, Case [15] concluded that history ”may be less
well served by classification and indexing than any other academic field” and
that the ”accomplished scholar - and particularly the historian - is not often
aided by the disciplinary boundaries that library classification schemes enforce.”
Instead of the ”disciplinary model of a body of knowledge, subdivided by place
and period”, the so called ”problem-oriented model” should be used as the basis
for the design of future tools and services for historians. At the same time, Case
correctly points out that it is not viable to fundamentally change documenta-
tion practices and reorder collections of archives and libraries but that special
services and tools might be able to bridge (semantic) gaps between the user and
existing knowledge bases.
3 Research Data
The Bundesarchiv is the Federal Archives of Germany who are responsible for
the permanent preservation and accessibility of federal archival documents from
the civil and military archives of the Federal Republic of Germany (since 1949)
and its predecessors. In addition, significant documents of private origins and
from political parties, associations and societies are kept in the archive. The
number of written inquiries to the Bundesarchiv amounts to roughly 60,000 per
5
Similar studies are, for example, from Collins [12], Conway [13], and Gagnon-Arguin
[14].
year, based on the numbers from 2008 and 2009.6 The Bundesarchiv has granted
supervised access to their user files which contain a physical documentation of
the correspondence and interaction between a user and the archive. Each user
file carries an identification number which is retrievable through a database
system offering a small range of search facets7 related to the user and associated
user files. Based on these facets a sample of 196 user files was retrieved, which
was further complemented by a special selection of 40 rich user files which had
been collected by the head of the department Stiftung Archiv der Parteien und
Massenorganisationen der DDR im Bundesarchiv (StA). The sample of user
files shares as a general historical and topical horizon Contemporary German
History (19th and 20th century) and contains rich and challenging inquiries. The
sampling process was informed by educated assumptions, professional advice of
archivists, and skimming through user files. The collection was stopped when
the questions extracted from the user files appeared to exhibit no new qualities
or substantial variances. Reliable information about the users’ background was
not available.
In total, 236 user files have been selected from which 100 were available for
further study. Only 60 user files contained at least one explicit or implicit infor-
mation request as part of an inquiry by email or letter. From these 60 user files,
546 single questions were extracted and pre-analyzed8 according to the method-
ology of Duff and Johnson [11] with very similar results. Regarding the type of
question, 260 questions were of type ”explicit” or ”implicit resource discovery”
(material-finding, specific form, specific item, consultation), 70 questions were
”factual”, and 216 questions consisted of ”administrative/directional”, ”user ed-
ucation”, or ”service request” questions. The questions of the type ”resource
discovery” and ”factual”, in total 330 questions, are part of the discovery stage
in the research process and are currently being analyzed as described in the
following sections.
4 Methodological Approach
The methodological approach taken in this study goes beyond the analysis of
the mere utterance level and syntactic structure of the inquiry and focuses on
the interpretation of the questions. Here, the sense of an inquiry is interpreted
in order to discover the implicit questions with regard to a certain domain of
discourse. In the scope of this work, two domains of discourse are being dis-
tinguished: the archival domain of record keeping and the domain of historical
6
http://www.bundesarchiv.de/oeffentlichkeitsarbeit/publikationen/taetigkeitsberichte/
7
This includes, for example, the general purpose of the inquiry as given by the user
on the user management form, a general subject and time frame of the inquiry’s
topic, or the department initially responsible for processing the inquiry which allows
concluding on the origins of the archival material. However, it is important to note
that these classifications are coarse and not meant for precise retrieval of user files
based on these search facets.
8
The publication of the results is in preparation.
inquiry for which traces and evidence can be expected to be found in the archive.
These two domains constitute the epistemological baseline for the interpretation
of the inquiries: What might the user need to know in order to satisfy his re-
search interest? Reality is then described in a way so that it fits the perceived
epistemological interest of the user and his question. This process is necessarily
an act of interpretation and relies on educated intuition regarding both domains
and necessarily filters probable implicit questions. It does not, to be sure, aim at
”truthful” models in terms of some perceived ”objective” meaning or structure
of a question. Regarding such epistemological issues of interpretation in relation
to historical science and theory of history, the approach to interpretation taken
here understands itself as meta-theoretical, similar to Gardin [16] in the domain
of archeology. It is agnostic to specific types of historical sciences but reflects
patterns applicable to general historical inquiry.
The patterns which are identified in the questions are modeled in CIDOC
CRM which describes historical facts in terms of possible relations between uni-
versals. It is the result of an empirical analysis of existing conceptualizations
of the cultural-historical world in the form of metadata structures. One of the
most important design principles of the CIDOC CRM is to represent the past as
discrete events. Material and immaterial persistent items are present at events
either as a concept or via a physical information carrier. History, therefore, is
conceptualized as meetings of persistent items through events in space-times
[17]. The historical-archival domain of the analyzed inquiries is in the scope of
the CIDOC CRM. For these reasons, its methodology is adopted in this work
and it is tried to identify whether the CIDOC CRM will completely or partially
cover the hypothetical ontology. In the latter case, appropriate extensions to the
CIDOC CRM will be proposed.
Formally, an ontology engineering approach is employed in that the inquiries
and their interpretations are being translated to an ontological model based on
the CIDOC CRM and appropriate extensions.
5 Translating Patterns of Questions to CIDOC CRM
Two examples will motivate how questions are being analyzed and how their in-
terpretation is formally represented in an ontological model based on the CIDOC
CRM. An inquiry typically consist of contextual information and one or more
direct or indirect questions.9
5.1 Example 1
Context: One source I would like to consult are the police- and surveillance re-
ports for the Weimar Republic which are about revolutionary movements. I would
9
All questions have been translated from German to English by the author of the
paper. Text in square brackets has been added either to make named entities anony-
mous or to clarify the meaning of certain paragraphs. Finally, red borderlines indicate
the entity at which a question is targeted.
like to know what the surveillance agency of the Reich (or the ones of the Länder)
had to say about [person name].
Question 1: Do you know if the Bundesarchiv holds such documents?
Question 2: Which agency of the Reich was responsible for the surveillance
of the revolutionary movements? The Reich or the Länder?
The given elements in these two questions and their context are the name of a
specific actor (”[person name]”), the type or function of a group (”revolutionary
movements”), the type or function of a legal body (”surveillance agency of the
Reich”), the type or function of documents (”police- and surveillance reports”),
and the name of a period (”Weimar Republic”).
The interpretation of the questions can be structured into two principle steps.
The first one is concerned with the wanted information asking for the research
interest of the user’s question: Which are probable or adequate answers to the
question with regard to the domain of historical inquiry but also to the archival
domain?
In the case of the first question the user is looking for reports which are the
result of a policing or surveillance activity targeted at a specific type of group
(”revolutionary movements”) or at a specific person (”[person name]”). In that
way, the first question could be even seen as a two-fold question. The results of
these policing or surveillance activities are documents about the activities of the
aforementioned actors. Such documents as routinely products of a governmental
institution are now stored in an archive. The user wants to know if such doc-
uments are available in the Bundesarchiv. Therefore, the information the user
wants are pointers to appropriate documents, for example, call numbers of files
likely to contain relevant documents.
The second question in the example is a fact-finding question. It operates
with the same given information but asks for a different wanted information.
The user wants to know which agency was generally responsible for surveillance
activities targeted at a specific type of actor. He is inquiring for a name of one or
more legal bodies. The word ”responsible” is important because it stresses the
fact that whatever agency conducted the surveillance activities did so following
a mandate which formally delegated said responsibility to the agency.
The second interpretation step comprises the translation of the question, its
context and its interpretation to the CIDOC CRM. The CIDOC CRM suggests
that historical facts and entities are related to each other through events which
form the world lines in history. Therefore, the second interpretation step asks
how the given and wanted information entities relate to each other.
The first two-fold question can be represented in CIDOC CRM as shown in
figure 1. This is a simplified representation expressing the formal basic struc-
ture of an answer adequate to satisfy the wanted information or the research
interest.10 The interpretation of the question is evident and materialized by the
10
The implicit question for pointers to documents, for example, a set of call numbers,
is not the point when translating to CIDOC CRM but the context of the documents
of interest. Identification for retrieving the actual physical document is not in the
scope of this ontological model.
documentation activity in the center of the figure. This class is a proposed ex-
tension to the CIDOC CRM and will be introduced in more detail later on. The
documentation activity is seen as being implicit in the historical reality referred
to in the question: The police- and surveillance reports have been created dur-
ing an event, or a series of events, which ”documented” some events which are
qualified by the participation of an actor (”[person name]”) or a specific type of
group (”revolutionary movements”). The documentation activity is following a
mandate which captures a specific type of ”documented plans (...) for deliberate
human activities”.
Most importantly, mandates specify or govern documentation activities. This
class is another proposed extension to the CIDOC CRM and will also be intro-
duced in more detail later on. In the case of the first two-fold question the
mandate has a specific type of group as its principle target and at the same
time aims at a specific actor. Furthermore, the mandate is assigned to an actor,
in this case an institution, who carries out the actual documentation activity
which, as the last relevant contextual information, falls within in the historical
period of the Weimar Republic. Documents which are the result of this constella-
tion are relevant documents and may adequately answer the user’s first two-fold
question.
Fig. 1. Question 1 in AKM
Figure 2 shows the translation of the interpretation of the second question to
the CIDOC CRM. An adequate answer can be modeled within the same pattern
as for the first question. In this case the wanted information is the name of an
actor who had the mandate to police or to keep under surveillance revolutionary
movements during the Weimar Republic.
These two questions and their representations in CIDOC CRM show a com-
mon core pattern which is grouped around a documentation activity which doc-
uments events and which is following a specific mandate. This relation between
documentation activity and mandate is essential. It can be identified in many
other questions through interpretation.
Fig. 2. Question 1 in AKM
5.2 Example 2
The second example shows that seemingly different questions exhibit very similar
patterns and that documentation activities based on mandates may cover a broad
range of different types of activities. Furthermore, some finer notions like self-
documentation and documentation of others are introduced in this example.
Context: In 1980, a delegation of the FDGB lead by Harry Tisch laid down a
wreath of flowers in Oradour. The visit was part of a trip of the FDGB to France
(demonstration in Limoges, reception and meeting with the FKP and CGT in
Paris). At this time, Tisch was also a member of the Politbüro of the ZK of the
SED.
Question 1: Where can documents be found about the planning [of this trip]...
Question 2: ...and the report on this trip?
Question 3: In your opinion, has such a trip been discussed or, at least, been
approved in the ZK?
The first question asks for documents about the planning of the trip to France
while the second question asks for the report on this trip. In both cases the
documents refer to the same event ”Trip to France” but they are the result of
two distinct activities. The first one, the planning activity, happens prior to the
actual trip and does not directly document the trip but series of planning events.
The second documentation activity, the reporting activity, produces one or more
documents which report on the trip event itself. Both questions ask for pointers
to documents as the result of their respective documentation activity.
Figure 3 combines the first and second question and their interpretations.
The documents are the result of documentation activities which document events
which were involved in the planning of the trip to France. In the case of the second
question, the documents are the result of a documentation activity reporting on
the event ”Trip to France”. Necessarily, both documentation activities followed
a mandate to do so and were carried about some actor.
Fig. 3. Questions 1 and 2 in AKM
The third question should adumbrate some more difficult issue in terms of
interpretation and translation of questions. The question asks if a specific actor,
the ”ZK”11 , had discussed or approved a specific event, the trip to France.
First of all, it is important to remember that the patterns are about the
general and generic relations between certain entities and not about the many
specific qualities of these connections: it is not relevant if the relation between
a document and an event is one of ”discusses” or ”approves” but that, on the
most generic semantic level, it is a relation of ”aboutness”. It is the genuine
task of the researcher to read and interpret the documents in order to find out
about the qualitative aspect if the ZK did in fact ”discuss” and, even more,
did ”approve” something. The pattern is a means for the researcher to discover
potentially relevant documents. One tentative inference which might be drawn
from a knowledge base which instantiates this ontological model is that the ZK,
or more precisely some members of this group, must have had knowledge of the
event ”Trip to France”.
Therefore, relevant documents for an adequate answer include those ones
which are about events during which the actor ZK was present and which in
some way refer to the event ”Trip to France”. An example for such an event
could be the planning event from the first question. Figure 4 shows another
possible scenario where the ZK carried out a committee meeting during which
the trip to France has been mentioned and which has been documented through
minutes.
Again, the minutes are the result of a documentation activity which follows a
mandate to take minutes. In this case, the ZK is the actor who not only follows
this mandate and carries out the documentation but also conducted the event
which is being documented. This is a kind of self-documentation as opposed
11
”ZK” is the abbreviation for Zentralkomittee (”central committee”) which belonged
to the Sozialistische Einheitspartei Deutschland (SED).
Fig. 4. Questions 3 in AKM
to the documentation of others in the case of surveillance and will be briefly
discussed in the next section.
6 The Documentation-Activity Pattern
The examples previously discussed exhibit a shared pattern which is able to
accommodate a broad range of different questions and their probable interpreta-
tion in terms of adequate answers. This section will introduce the current draft
of the Documentation-Activity pattern as shown in figure 5. So far, this pattern
appears to be one of the most prominent and complex results from the analysis
of the user questions.12
At the core of this pattern resides a new proposed class E7.1 Documentation
Activity. This new event class is an extension to the CIDOC CRM in order to
appropriately capture the essentials of activities which, literally speaking, docu-
ment E5 Events and which create one or more E31 Documents. It is a sub-class
of E65 Creation and not of E7 Activity because a characteristic feature of the
documentation activity is the creation of documents and only events of the type
E65 Creation ”result in the creation of conceptual items or immaterial products”
through P94 has created. Furthermore, the scope of E7.1 Documentation Activ-
ity is more specific than that of E65 Creation in that documentation activities
document E5 Events and, most importantly, follow a mandate. The represen-
tation of the fact that a documentation activity follows a mandate led to the
introduction of a new property called follows mandate.
The E29.1 Mandate is the second proposed extension to the CIDOC CRM
as a sub-class of E29 Design or Procedure. The mandate formulates the principle
scope of application for documentation activities in that it specifies who has the
mandate to execute the documentation activity and which specific actors, types
12
While the analysis of the questions is on-going and no reliable evidence based on the
current research sample can be provided at this point, an estimate of at least 30%
of all questions in the sample might be adequately covered by this pattern either
partially or in full.
Fig. 5. The current draft of the Documentation-Activity pattern
of actors, or types of activities may be the target. In order to appropriately
describe these target relations new properties – target function, target group,
target actor, and has mandate – have been introduced.
The E7.1 Documentation Activity and the E29.1 Mandate related through
follows mandate constitute the essential core of the identified common pattern:
The documentation of events according to standing mandates producing docu-
ments which can be found in the archive. This mandate-based documentation
(auftragsgemäße Dokumentation) can not be adequately represented with E65
Creation and E29 Design or Procedure. The pattern allows to draw conclusions
on the probability that specific types of events have been documented and that
traces can be expected in the archive.
The documentation activity and its contextual classes can be seen as being
part of a description of the historical reality as given in the user’s question.
The mandate, on the other hand, belongs to an intentional level (Absichtsebene)
where principles are formulated which are meant to formally govern the historical
reality and which might find their expression in documents. These documents
are the point where this ontological representation of the historical reality would
intersect with the one of the archival domain of record keeping as indicated in
figure 5. It is important to note that an E31 Document is not a physical item but
”comprises identifiable immaterial items that make propositions about reality”.
A physical materialization of an E31 Document in the archive may be an E33
Linguistic Object which ”comprises identifiable expressions in natural language
or languages”. Here, a model of expressions of documents in the archive is not
included.13
13
Cf. [18] for an approach to mapping EAD to the CIDOC CRM.
The analysis of the questions is on-going and changes to the pattern might
occur and there are other aspects which appear to be relevant. The official and
unofficial nature of a document, for example, seems to be another important
aspect. This point cannot be discussed in any detail in this paper, however, if a
document is official or unofficial is most likely determined by the circumstances
of its publication. As already mentioned, the examples also show cases in which
the documentation activity is carried out by the same actor who is also respon-
sible for the documented activity. This is a kind of self-documentation giving
an ”official account” (Rechenschaftsbericht) such as proceedings, government
statements etc. In the Documentation Activity pattern this can be expressed by
two principle sub-types E7.1a Self-Documentation and E7.1b Documentation of
Others.
7 Conclusion
This paper introduced the draft of the Documentation-Activity pattern which
is part of the Archival Knowledge Model (AKM). The AKM is an ontological
model which comprises representations of general patterns found in archival user
inquiries and their interpretations.
Such an ontological model can help to bridge the semantic gap between tra-
ditional archival documentation and organizing principles and the conceptual-
izations employed by different kinds of users and support building search and
discovery systems which are able to better respond to pattern-oriented questions.
As a formal model, the AKM could also inform the design of archival metadata
schemas or new archival ”cataloging rules” as, for example, that titles of series
or files should not be plain text but structured according to patterns like the
Documentation-Activity pattern.
References
1. Unsworth, J.: Scholarly primitives: What methods do humanities researchers have
in common, and how might our tools reflect this? (2000)
2. Menne-Haritz, A.: Access: The reformulation of an archival paradigm. Archival
Science 1 (2001)
3. Cox, R.: Revisiting the archival finding aid. Journal of Archival Organization 5(4)
(2008)
4. Craig, B.: Perimeters withfences?or thresholds with doors? two views of a border.
American Archivist 66(1) (2003)
5. Sinn, D.: Room for archives? use of archival materials in no gun ri research.
Archival Science 10(2) (2010)
6. Anderson, I.G.: Are you being served? historians and the search for primary
sources. Archivaria (58) (2004)
7. Doerr, M., Iorizzo, D.: The dream of a global knowledge network: A new approach.
Journal on Computing and Cultural Heritage 1(1) (2008)
8. Dworman, G.O., Kimbrough, S.O., Patch, C.: On pattern-directed search of
archives and collections. Journal of the American Society for Information Science
51(1) (2000)
9. Shaw, R.: Information organization and the philosophy of history. Journal of the
American Society for Information Science and Technology 64(6) (2013)
10. Harris, C.: Archives users in the digital era: A review of current research trends.
Dalhousie Journal of Interdisciplinary Management 1 (2005)
11. Duff, W.M., Johnson, C.A.: A virtual expression of need: An analysis of e-mail
reference questions. American Archivist 64(1) (2001) 43–60
12. Collins, K.: Providing subject access to images: A study of user queries. American
Archivist 61(1) (1998)
13. Conway, P.: Partners in Research: Improving Access to the Nation’s Archive.
Archives & Museum Informatics, Pittsburgh (1994)
14. Gagnon-Arguin, L.: Les questions de recherche comme matriau dtudes des usagers
en vue du traitement des archives. Archivaria 46(1) (1998)
15. Case, D.O.: The collection and use of information by some american historians: A
study of motives and methods. The Library Quarterly 61(1) (1991)
16. Gardin, J.C.: Archaeological discourse, conceptual modelling and digitalisation:
An interim report of the logicist program. The Digital Heritage of Archaeology:
Computer Applications and Quantitative Methods in Archaeology, Proceedings of
the 30th Conference, Heraklion, Crete, April 2002, CAA 2002 (2002)
17. Doerr, M.: The CIDOC conceptual reference module: An ontological approach to
semantic interoperability of metadata. AI Magazine 24(3) (2003)
18. Bountouri, L., Gergatsoulis, M.: Mapping encoded archival description to CIDOC
CRM. Proceedings of the First Workshop on Digital Information Management
(2011)