Adding Semantic Extension to Wikis for Enhancing Cultural Heritage Applications

TL;DR: It is highlighted how wikis can be relevant solutions for building cooperative applications in domains characterized by a rapid evolution of knowledge, and the capabilities of semantic extension to provide better quality of content, to improve searching, to support complex queries and finally to carry out different type of users.

Abstract: Wikis are appropriate systems for community-authored content. In the past few years, they show that are particularly suitable for collaborative works in cultural heritage. In this paper, we highlight how wikis can be relevant solutions for building cooperative applications in domains characterized by a rapid evolution of knowledge. We will point out the capabilities of semantic extension to provide better quality of content, to improve searching, to support complex queries and finally to carry out different type of users. We describe the CARE project and explain the conceptual modeling approach. We detail the architecture of WikiBridge, a semantic wiki which allows simple, n-ary and recursive annotations as well as consistency checking. A specific section is dedicated to the ontology design which is the compulsory foundational knowledge for the application.

Summary

1 Introduction

• Collaborative platforms that manage scientific knowledge are essential tools for scientists to help them to formalize their ideas, to develop theories collaboratively, to publish results (research articles, technical reports, data sets, etc.) and to produce knowledge for different kind of users.
• Most of wikis also provides a versioning system to track content changes and a full-text search engine for querying wiki pages.
• Nevertheless, a mere document management system is not sufficient to catch interdependent structures of knowledge.
• Domain specialists often need to comment on primary data.
• Adding semantics yields two dimensions of enhancements to a wiki: 1) adding a more formal structure to the wiki; 2) exporting, integrating and reusing information by the adoption of standard semantic technologies.

2 Project overview

• The aims of the international project CARE (Corpus Architecturae Religiosae Europeae) is the setting up of a corpus describing Christian edifices in Europe (http://care.u-bourgogne.fr).
• Italy, Spain, Czech Republic, Poland, Slovakia, France and Croatia have been included in the project four years ago.
• The French corpus focuses on the 7th and 8th centuries with very rich decades in terms of number of monuments.
• The French part of the project is supported by a French ANR funding (ANR-07-CORP-011).
• Building collaborative platform brings out some challenging characteristics: 1) complexity of data (heterogeneous, incom- plete, uncertain, inconsistent, spatial, temporal); 2) domain knowledge barrier; 3) evolving knowledge; and 4) skills of actors.

2.1 The CARE community

• From an organizational perspective, the CARE project takes the form of an expert network collecting and providing information on edifices, analyzing historical sources, filling documents and collaborating in the exploitation of the corpus through smaller research groups.
• Furthermore, the project involves undergraduate students that help in collecting information but lack expertise required to interpret data.
• In France, more than sixty researchers are collecting and analyzing data concerning approximately 2700 monuments.
• Two key characteristics describe the CARE community: multi-disciplinarity:.
• It is designed as a collaborative process which merges information from various disciplines; inter-disciplinarity:.

2.2 Conceptual modeling for the foundational knowledge

• Linster in [11] shows that the interaction among domain experts, knowledge engineers and tools creates the knowledge.
• He has shown that the process of elaborating a knowledge-based system is a constructive model-building process that includes: a discussion process between knowledge engineers and domain experts as well as the construction of a conceptual model (i.e. a general and abstract framework).
• The authors have applied Linster’s guidelines to the CARE project in order to initiate a foundational knowledge from the corpus of documents.
• All changes of space or religious function determine a new edifice state.
• Salient concepts let us to build a conceptual model, three groups of elements have been identified : 1. spatial concepts without temporal relationship (light grey): concept EGS refers to edifice, group of edifices or space inside edifice such as nave or apse.

3 WikiBridge’s architecture

• In a survey authored by Uren et al. [15], authors study semantic annotation, identify a number of requirements, and review some semantic annotation systems.
• WikiBridge’s design principles, in agreement with the CARE community, are following of the seven requirements given by Uren et al.: easy to use interface, user collaborative design, support of different user skills, support of heterogeneous format, compatibility with Semantic Web standards, annotation capabilities and storage, support for reasoning.
• In the next subsection the authors develop the most important requirements with regards to the architectural design of WikiBridge.

3.1 Requirements

• In a knowledge engineering process, it is common that non-technical domain experts work together with experienced knowledge engineers.
• To support different levels of users skill certain advanced functionalities should be hidden from novice users but made available to experienced users.
• Thus, the authors use an Access Control List (ACL) mechanism to describe privilege control depending on user identity and group affiliation.
• Advanced users can define forms to help users to enter descriptions of edifices, each part of a form generates automatically annotations.
• To be able to exchange data with other applications (e.g. ontology editors, Web Services, other wikis), a compliance to Semantic Web standards is required.

3.2 Architecture

• One of the most famous semantic wiki is Semantic MediaWiki (SMW) which is based on MediaWiki [16, 9].
• In 2009, when the authors have started the project, complex annotation and consistency checking were identified as mandatory functionalities.
• The authors have extended MediaWiki with the following semantic components: form based acquisition interface with automatic annotation, annotations wizard, annotations validation based on the context of a document, semantic rules and a semantic query engine.
• The annotation wizard helps users to construct simple or complex annotations by selecting ontology terms in lists and giving them properties and values .
• Two kinds of constraints can be check by using the ontology knowledge: 1) domain values of properties using ABox capabilities; and 2) structural consistency of properties using TBox capabilities (for instance, a cathedral can have a nave but cannot have an atrium).

3.3 Users with different skills

• Information access has been designed with taking into account some features about users.
• The authors have thus identified a usage typology in accordance to 1) kind of usage (reader, investigator, annotator); 2) knowledge degree of the domain (domain specialists like historian researchers and non specialists).
• Nevertheless all types of queries rely on the SPARQL query engine which also allows to process in line queries into wiki pages in order to summarize informations.
• To operate spatial and temporal analysis on annotations a set of web services has been developed .

4 Offering semantic tools for archaeology

• Cultural heritage collections can be annotated with different thesauri.
• //tinyurl.com/5u8bjer maintains about twenty thesauri, classified by content and organization, also known as The web page http.
• CIDOC-CRM aims at treating all types of material collected and/or displayed by museums: sites, monuments as well as collections of fine and applied arts.
• It covers also contextual information: historical, geographical context in which materials are placed and which gives them much of their significance and value.

4.1 Religious concepts in the CARE project

• Religious concepts in CARE are edifices, represented by the concept Building, with its decomposition into different constituent elements (nave, transept, apse, etc.) represented by the concept of Structure.
• To detail parts of an edifice, the authors introduced the concept of ElementArchitectural to describe masonry, floor, opening, etc. Liturgical installations (altar, ambo, ciborium, etc.) are represented by the concept of InstallationLiturgique and burials represented by the concept Tombe.
• These concepts have been placed under the concept E24 Physical Man-Made Thing CIDOC-CRM.
• Indeed, CIDOC-CRM defines this concept as ”all persistent physical items that are purposely created by human activity”.
• Figure 6 represents all these concepts (concepts with EXX are CIDOC-CRM concepts).

4.2 Modeling spatial relationships in archaeology

• The geometry implementation in the textual descriptions that are analyzed is a complex geometry.
• Indeed, these descriptions do not refer to an absolute and orthonormal space: it is rather, a space perception or a cognitive space whose structure is largely based on the functional aspects and objects described, and the perspective of the archaeologist.
• To represent the topological properties, the authors used the work of Hegenhofer and Herring [8].
• The authors have defined a minimum set of eight relations (disconnected, externally connected, partially overlap, equal, tangential proper part, non-tangential part, tangential proper part inverse, non-tangential part inverse) describing the relations between two regions.

4.3 Modeling temporal knowledge to track evolutions in CARE project

• When writing of his excavation report, the archaeologist graphically summarizes the results obtained with a timeline which is often organized by anterior/posterior relationships: materials are considered in relation to each other.
• In the CARE project, time model is based on following criteria: some absolute benchmarks and a relative chronology based on intervals.
• These century divisions are placed under E52 TimeSpan.
• Some Allen’ relationships [1] are properties used.
• This notion includes both complex and long-lasting actions such as building an edifice, as well as simple and short-lived actions.

5 Related works

• Several semantic wikis have been developed or used specifically for cultural heritage applications.
• Authors have implemented the ideas for the German Handbuch der Architektur, a comprehensive multi-volume encyclopedia of architecture.
• The wiki consists of three parts: a collection about twenty essays giving a comprehensive domain walk-through, translations of the describing ancient sources and a glossary.
• The framework integrates a wiki engine for rendering documents stored in the ontological tier.
• MANTIC integrates different data sources and the global schema is based on CIDOC-CRM.

6 Conclusion

• The authors have presented WikiBridge which provides users with advanced functionalities such as rich annotation model, consistency checking.
• The first experiment of the use of WikiBridge shows many interesting possibilities for scientific community, mainly the possibility given to scientists share and collaboratively build annotated knowledge.
• The authors have demonstrated that flexibility and data quality required by scientific applications can be achieved by using wiki with Semantic Web technologies.
• The semantics of annotation is guaranteed by an ontology including constraints which allow to describe accurately domain knowledge.
• The authors dual approach allows to cope with evolution of knowledge by dynamically modifying the ontology and annotations without modifying database schema.

Figures

Fig. 2. WikiBridge’s Architecture

Fig. 3. User interaction layer

Fig. 5. OpenLayers interface interacting with a web service of WikiBridge

Fig. 4. Annotation wizard

Fig. 6. Religious concepts in CARE project

Fig. 1. Conceptual model of the CARE corpus

Full text

HAL Id: hal-00934969
https://hal-univ-bourgogne.archives-ouvertes.fr/hal-00934969
Submitted on 22 Jan 2014
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Adding Semantic Extension to Wikis for Enhancing
Cultural Heritage Applications
Eric Leclercq, Marinette Savonnet
To cite this version:
Eric Leclercq, Marinette Savonnet. Adding Semantic Extension to Wikis for Enhancing Cultural Her-
itage Applications. International Conference on Digital Information and Communication Technology
and its APplications, 2011, France. pp.348-361. �hal-00934969�

Adding Semantic Extension to Wikis for
Enhancing Cultural Heritage Applications
´
Eric Leclercq and Marinette Savonnet
LE2I Laboratory UMR CNRS 5158
Universit´e de Bourgogne
9, Avenue Alain Savary
21078 Dijon, France
{Eric.Leclercq,Marinette.Savonnet}@u-bourgogne.fr
http://le2i.cnrs.fr
Abstract. Wikis are appropriate systems for community -a u t h o red con-
tent. In the past few years, they show that are particularly suitable for
collaborative works in cultural heritage. In this paper, we highlight how
wikis can be relevant solu ti o n s fo r b u i ld i n g cooperative applicati o n s i n
domains characterized by a rapid evolution of knowledge. We will po int
out the capabilities of semantic extension to provide better quality of
content, to improve searching, to support complex queries and finally
to carry out different type of users. We describe the CARE project and
explain the conceptual modeling approach. We detail the architecture of
WikiBridge, a semantic wiki which allows simple, n-ary and recursive an-
notations as well as consistency checking. A specific section is dedicated
to the ontology design which is the compulsory foundationa l knowledge
for the application.
Key words: Semantic Wiki, Ontology Engine erin g , Cultural Heritage
application
1 Introduction
Collaborative platforms that manage scientific knowledge are essential tools for
scientists to hel p them to formalize their ideas, to develop theories collabora-
tively, to publish results (research articles, technical reports, data sets, etc.) and
to produce knowledge for different kind of users. Moreover, a collabor ati ve plat-
form should be able to integrate oth er services such as v i su al iz at ion tools, or
spatial analysis tools.
Wiki sol u t i ons meet th e requirements of a web platf or m with c ollaborative
capabilities. Easy setup and ri ch editing support are pri mar y reasons for the
widespread adoption of wikis. Users can enter text and others types of data
(pictures, video) and connect content through hyperlinks. Most of wikis also
provides a versioning system to track content changes and a full-text search
engine for querying wiki pages.

The narrative structure is one advantage of wiki documents centric approach,
compared to a database centric approach. In a database centric approach, the
database schema is built upon entities identified in the first step of analysis,
and thus based on an instant knowledge. In domains characterized by a rapid
evolution of knowledge, such as biol ogy or archaeology, a static dat abas e schema
is not suitable and can be proscribed by the cost of evolution. Nevertheless, a
mere document managem ent system is not sufficient to catch interdependent
structures of knowledge. For example, domain specialists often need to comment
on primary data. Adding semantic annotation capabilities to documents allows
different levels of interpretation and can sustain: 1) knowledge evolution by keep-
ing track of the successive annotations; 2) better quality in the query evaluation
process; and 3) amenable result displayed according to user skills. Annotations
can be defined at a coarse grained level (wh ole document) or at a fine grained
level (i.e. attached to a piece of text). An ontology must be associated to the
annotation system to provide a semantics for annotation terms according to
domain knowledge.
Semantic wi k i solutions meet the requirements of annotation system and
knowledge description. Adding semantics yields two dimensions of enhancements
to a wiki: 1) adding a more formal structure to the wiki; 2) exporting, integrat-
ing and reusing information by the adop t ion of standard semantic technologies.
Compared to a traditional database, a semantic wiki allows: 1) to expand the
structure of documents content; 2) to enable a data model emergence from the
usage; and 3) to support collaborative, distributed workflows and processes. Se-
mantic wiki thus seems to combine the best from two worlds: structure f r om
databases as well as expandability and collaboration capabilities from wiki sys-
tems.
The re st of the paper is organized as follow: section 2 gives an overview of the
CARE project, section 3 describes the requirements and WikiBridge architec-
ture, section 4 describes semantic tools for archaeology, and section 5 discusses
related works. Finally, section 6 concludes the paper.
2 Project overview
The aims of the international project CARE (Corpus Architecturae Religios ae
Europeae) is the setting up of a corpus describ i ng Christian edifices in Eu-
rope (http://care.u-bourgogne.fr). Italy, Spain, Czech Republic, Poland,
Slovakia, France and Croatia have been included in the project four years ago.
Each edifice is described in a document that focuses on the description of stat es
of evolutions from the 4th century to the 11th century. The French corpus fo-
cuses on the 7th an d 8th centuries with very rich decades in t er ms of number
of monuments. The French p art of the project is supported by a French ANR
funding (ANR-07-CORP-011).
Representing and managing knowledge in cultural heritage requ i re a deep
understanding of specific concepts. Building coll aborative platform brings out
some challenging characteristics: 1) complexity of data (heterogeneou s, incom-

plete, uncertain, inconsistent, spatial, temporal); 2) domain knowledge barrier;
3) evolving kn owledge; and 4) skills of actors.
2.1 The CARE community
From an organizational perspective, the CARE project takes the form of an ex-
pert network collecting and providing information on edific es , analyzing histori-
cal sources, filling documents and collaborating in the exploitation of the corpus
through smaller research groups. Furthermore, t h e project involves undergrad-
uate students that help in collecting information but lack expertise required to
interpret data. In France, more than sixty researchers are collecting and ana-
lyzing data concerning approximately 2700 monuments. Two key characteristics
describe the CARE community:
multi-disciplinarity: The dat a collecting process involves archaeologists, his-
torians, art historians, topographers, draftsmen. It is designed as a collabo-
rative process which merges information from various discipli ne s;
inter-disciplinarity: The interpretation of data brings together all th e actors
which also enrich their respective practices by the confrontation of methods
or prob l ems .
2.2 Conceptual modeling for the f ou n dat ion al knowledge
Linster in [11] shows that the interaction among domain experts, knowledge
engineers and tools creates the k n owledge. He has shown that the process of
elaborating a knowledge-based system is a constructive model-building process
that includes: a discussion process between knowledge engineers and domain
experts as well as the construction of a conceptual model (i.e. a gener al and
abstract framework). Thus, the knowledge engineering act i v ity encompasses the
design of two kinds of models: model to make sense and model to implement
systems.
We have applied Linster’s guideline s to the CARE project in order to ini-
tiate a foundational knowledge from the corpus of documents. The first stage
is the salient concepts identification. The key concept is the edifice to whi ch it
is essential to model changes. All constituent elements of a building need to be
described. They can delimit space or define religious function (baptismal , fu-
nerary, etc.). All changes of space or religious function determine a new edifice
state. Edifices and their evolutions are described in a set of documents. Salient
concepts let us to build a conceptual model, three groups of elements have been
identified (figure 1):
1. spatial concepts without temporal relationship (light grey): concept EGS
refers to edifice, group of edifices or space inside edifice such as nave or apse.
Composition relationships can be identified between EGS.
2. spatio-temporal concepts (grey) called SEGS, represent variations of spatial
concepts in time. Spatio-temporal concepts are linked to a date or a period.
Dating elements can be determined by documents, or described by methods
such as C14, ther mol u mi ne sc en ce or stratigr ap hy.

3. identification of vocabulary terms for the project domain (black). While
concepts EGS and SEGS are used to structure the descriptions, terms are
used to describe specific elements, properties of edifices, religious functions
or manufacturing techniques.
Dating element
Type: String
Location: String
Function: String
EdificeGroupSpace(EGS)
0..*
1..*
Document
Title: String
well−documented 1..*0..*
DatingElement
0..*
0..*
attests
StateEdificeGroupSpace(SEGS)
Type: String
Function: String
1..*1 links
1..*
0,1
has
constituent element
Religious functionEdifice space (nave, transept, altar, burial, ...)
Fig. 1. Conceptual model of the CARE corpus
The des cr i pt i on of the construct i on of the ontology is detailed in Section 4.
3 WikiBridge’s architecture
In a survey authored by Uren et al. [15], authors study semantic annotation,
identify a number of requirements, and review some semantic annotation sys-
tems. WikiBrid ge’ s design principles, in agreement with the CARE community,
are following of the seven req u ire m ents given by Uren et al.: easy to use in-
terface, user collaborative design, support of different u se r skills, support of
heterogeneous format, compatibility with Semantic Web standards, annotation
capabilities and storage, support for reasoning.
In the nex t subsect i on we develop the most important r eq ui r em ents with
regards to t h e architectural design of WikiBridge.
3.1 Requirements
In a knowledge engineering process, it is common that non -t e chnical domain ex-
perts work together with experienced knowledge engineers. To support different
levels of users skill certain advanced functionaliti e s should be hidden from novice
users but made available to experienced users. Thus, we use an Access Control