Ontology Modification in a Multi-User Concept
Conflict Resolution
Fatma Chamekh, Guilaine Talens and Danielle Boulanger
Magellan - IAE- Jean Moulin University, 6 cours Albert Thomas – BP 8242, Lyon, France
Keywords: Ontology Modification, Multi-Agents System, Multi-User Context, Conflict Situation.
Abstract: The task of ontology evolution has been a topic of several research works leading to a number of tools
facilitating the process of ontology evolution. These tools often propose different approaches without taking
into account the multi-user context. It is either assumed that an ontology engineer changes the ontology, or
that experts modify the ontology in asynchronous way. In this paper, we propose an approach to update
ontologies for keeping knowledge up to date. This approach is based on agent’s paradigm. The relevance of
agents is that they can interact with each other to assist the expert to upgrade dynamically knowledge. In
multi-user context, experts are able to modify simultaneously the same ontological entity which can
generate conflict situations in the system. This one could generate ontology inconsistency. To control the
conflict situation each agent uses predefined rules.
1 INTRODUCTION
In the last decade, the information technology
explosion has led to changes in the organization and
management of companies. In order to improve their
reaction and adaptation period, companies have
interested in tools supporting capitalization and
management (Gandon, 2002). Knowledge
management systems have helped organizations to
capture, to capitalize and to share their knowledge.
They enrich their in-house knowledge by capturing
statements extracted from various sources of
information (documents, databases, websites...).
These statements could be transformed into ontology
compatible elements or axioms, to describe domains
in a formal way. However, the dynamic facet of
domain has brought about ontology evolution to
keep the knowledge up to date.
Handling the ontology evolution process in a
distributed, decentralized environment is time-
consuming and complex process. Therefore, we
propose a framework to modify the ontology from
document using the agent’s paradigm to reduce the
interventions of experts
Using multi-agent system (MAS) is motivated by
the following reasons: i) The MAS runs when a new
document is added, modified or deleted. Agents use
its reactive behaviours to begin this incremental
process. ii) To keep ontology consistency, agents use
its local knowledge. iii) Interactions between agents
support the ontology evolution process.
In a multi-user environment, experts can
simultaneously modify the ontology. Collaborators
change ontology in asynchronous mode, each one
modifies the same version without knowing the
modification made by the other one, which leads to
conflict situation. Conflict is natural phenomenon in
every field of human word, hence it express the
divergence of interests or needs. Because the solving
of ontology evolution’s problem is divided among a
number of agents, the conflicts between them
happens in the case of incompatible goals.
Therefore, it makes sense to examine the conflict
situation, with the aim to achieve the ontology
modification well.
In traditional way, collaborative conflicts are
usually handled by mechanisms of locking or
branching/merging provided by version management
system (eg, cvs, subversion, Git, etc) (Chen et al.,
2009). In this case, a conflict occurs when two
developers modify the same source file. However, in
the case of ontology evolution, users modify
concepts and relationships between them. So, the use
of the mechanisms listed is not possible.
In this paper, we try to answer the research
questions: How design a multi-agent system to
support ontology evolution in a multi-user context?
This research question breaks down into the
296
Chamekh F., Talens G. and Boulanger D..
Ontology Modification in a Multi-User Concept - Conflict Resolution.
DOI: 10.5220/0005081602960303
In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2014), pages 296-303
ISBN: 978-989-758-049-9
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
following sub-questions: (i) how to guide the expert
to evolve ontology by keeping its consistency? (ii)
How to define and manage the conflict situations?
The reminder of this paper is organized as
follows: the section 2 presents some existing
ontology evolution approaches. In the section 3 we
discuss a use case scenario behind this work. In
section 4 we present an overview of our framework.
In section 5 we show our methodology for keeping
consistency and resolving conflicts in the context of
ontology evolution process supported by multi-agent
system. Future works and conclusion are presented
in section 6.
2 RELATED WORK
According to (Stojanovic, 2004), ontology evolution
is defined as the “timely adaptation of ontology to
the arisen changes and the consistent propagation of
these changes to dependent artefacts”.
Many ontology evolution frameworks rely on
inconsistencies control lists that define the
consequences of each change. To resolve
inconsistencies of KAON (KArlsruhe ONtology)
ontology, (Klein, 2004) propose a set of
preconditions and post-conditions. Consistology
(Jasiri et al., 2010) is a tool where providers
ontology consistency using change kits (a set of
rules and suggestions) which control the
inconsistencies generated by each type of change.
Other approaches specify the consistency
between components of the semantic web. Among
others, (Luong et al., 2006) present CoSWEM
(Corporate Semantic Web Evolution Management)
as a system to manage effects of the ontological
changes to the semantic annotations. They define
rule-based approach for solving inconsistency, and
(Rogozan et al., 2005) propose an approach for
ontology evolution relatively to educational
semantic web. They implement the Semantic
Annotation Modifier system to keep consistency
between ontology and resources.
Other frameworks introduce the patterns for
maintaining overall consistency. Onto-Evoal
(Ontology Evolution-Evaluation) (Djedid et al.,
2010) present inconsistency resolution patterns.
EVOLVA (Zablith et al., 2014), is an ontology
management framework. It explores background
knowledge sources (Wikipedia, WordNet, online
ontologies….). The authors use a pattern-based
approach to verify the relevance of the change
against the base ontology.
Ontology evolution may consider also the multi-
user context. Experts could modify ontology
simultaneously. (Noy et al., 2006) propose two
plugins: the change management plugin and the
PROMPT plugin, integrated in PROTEGE to
support the ontology evolution in the collaborative
environment. The system use CHange and
Annotation Ontology (CHAO) to describe the
changes between versions. Each user annotates the
change made.
There are few approaches investigating the
problem of ontology evolution coupled with MAS.
DYNAMO is adaptive multi-agent system (AMAS)
for building and evolution of Terminological and
Ontological Resource (TOR) from texts (Sellami et
al., 2012). Each term and each concept try to find its
right place in the AMAS organization that is the
ontology. A set of behaviours for each type of agent
is defined. Local rules are adopted to detect non-
cooperative situation and actions to be taken to go
back in a cooperative state. The ontology engineer
uses inconsistence sheets which contain the
inconsistency code and the changed term, to reach
the modification. Another study of ontology
evolution and MAS is given in (Rahman et al,.
2012). The authors present MAEKM (Multi Agent
Enterprise Knowledge Management) based on
ontologies modelling functional domains and multi-
agent architecture performing the data retrieval and
managing the changes that may occur within the
data sources.
Our work presented in this paper can be
compared with some similar existing studies.
(Sellami et al., 2012) has presented DYNAMO as
MAS for building and evolution TOR from texts.
Nevertheless, this approach considers the agent as
term and the MAS as ontology and it not has been
mentioned the ontology evolution process. In
(Rahman et al., 2012), the MAS manage the
ontology consistency. However, they do not use
predefined rules to solve the consistency and the
collaborative context did not dealt with. Our
evolution management system designs the ontology
evolution process from documents based on agent’s
paradigm. Our work differs from the MAS system in
DYNAMO for assigning each step of the process as
agent’s role. Relying on rule-based approach, our
framework, encapsulate these rules as agent
knowledge base to reach with inconsistency.
Regarding to ontology evolution in a multi-user
context, .(Noy et al., 2006) has presented CHange
and Annotation Ontology. Despite that, this
approach only presented a way to annotate the
OntologyModificationinaMulti-UserConcept-ConflictResolution
297
change of each user but they do not specify
techniques to verify and solve it.
3 USE CASE SCENARIO
As an example to illustrate concepts discussed so
far, we propose an ontology describing aspects
related to a healthcare domain especially general
medicine, which acts as information based from
Ontologos corporate (www.ontologos-corp.com).
This information base contains medical reports and
ontology. In order to describe the general medicine
domain, the doctors define terms and relations.
The medical report is a document in which the
doctor registers a clinical case study. This one is
related to research work, for example the report
number83 explains the role of the obesity in growing
the asthma.
The first state of the system is to implement a
semantic research engine. Due to the increase of the
medical report, the existent ontology could be
changed. Further to this, the expert needs a system to
guide him to modify the ontology. Given this, the
system requires to capture these changes for it to
efficiently serve the expert and reduce
inconsistencies.
Suppose that users add a new document. The
system may analyse the expert’s request in order to
identify the change that could be done to the
ontology (add concept, add instance, add property).
Once we have the type of modification, the system
analyses the change by studying the effect on the
consistency of ontology and between ontologies and
documents. To identify the adequate operations
related to each type of change, it is necessary to
determine the type of change and inconsistencies. If
different possibilities exist, i.e., different additional
operations can be applied with different effects, the
users have to choose the appropriate additional
changes to implement. Various operations are
displayed to experts in order to assist them. Experts
validate the operation of change. The system checks
the consistency of the dependent artefacts
(application, document) after each change.
At the end, changes are merged from different
experts. The system checks the coherence of the
modifications by using a set of conflict resolution.
When the system reaches the consensus, the new
version of ontology (Vn+1) is created. The old
version is saved in the log of version and all changes
are backed up in the log of changes.
4 PLATFORM ARCHITECTURE
Our approach is based on the ontology evolution
process identified by (Stojanovic, 2004) and (Klein,
2004). This process is made up of four steps:
identification of changes, analysis of changes, and
propagation of changes and management of
versions. We assign to each agent a process step.
Agents interact to run different processing steps.
The architecture of the proposed framework is
shown in Figure 1. The system is composed of four
agents: the user agent, the ontology agent, the
consistency agent and the version agent, and three
components: The Learning module, the log of
changes and the log of versions (Chamekh et al.,
2013).
We focus our ontology model to the OWL DL
which is an axiom-oriented language. Classes and
properties have structural descriptions specified
through some defined constructors (Horrocks et al.,
2003). Satisfying ontology within an interpretation is
constraint by the satisfaction of all ontology axioms.
For convenience we will adhere to more compact,
traditional SHOIN(D) syntax. For the
correspondence between this notation and various
OWL DL syntaxes see (Horrocks et al., 2004).
Definition 1: an ontology O is a set of concepts,
properties and individuals.
O={C, P, I}
C= {c
1
,c
2
,c
3
,....c
n-1
,c
n
}
P= {p
1
,p
2
,p
3
,....p
n-1
,p
n
}
Example 1: As a running example, we will consider
an extract of general medicine domain, consisting of
the following axioms:
Non inflammatory pathology Pathology,
inflammatory pathology Pathology, (Non
inflammatory pathology and inflammatory
pathology are pathologies)
Infection inflammatory pathology Non
inflammatory pathology
4.1 User Agent
It detects the change to be realized by analysing the
user request. In order to perform all tasks related to
the management expert request, we have provided
several behaviours: analyse the user request,
communicate with the other agents and user, trigger
learning module and verify the existence of
documents.
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
298
Figure 1: Platform architecture.
When the user agent analyses the request, three
alternative operations exist : Add document, Delete
document, Modify document:
4.2 Learning Module
This module receives a document as input. Using
Text2Onto (Maedche, et al,2000) which is a
framework for data-driven change discovery by
incremental ontology learning. It uses natural
language processing and text mining techniques. The
output generated is XML/OWL file. This one
contains the extracted terms. These terms are
categorised as concepts, instances and taxonomic
relations.
As example of concept presentation:
<a:Concept rdf:ID="non infection inflamatory pathology_c">
<a:Ratingrdf:datatype="http://www.w3.org/2001/XMLSchema#d
ouble">0.0136986301369863</a:Rating>
<owlx:Labelrdf:datatype="http://www.w3.org/2001/XMLSchema
#string">night</owlx:Label> </a:Concept>
Example2. Let’s consider a change made by user 1
which adds a new medical report. The user agent
verifies if the document exists. If it is true, the user
agent informs the user that the document exists and
he must drop this operation. If it not, the user agent
triggers the learning module, and extracts:
The concept: Non infection inflammatory pathology.
4.3 Ontology Agent
It analyses a change. In order to perform all tasks
related to the management of the user request, we
provide this agent by various behaviours: detect the
ontology to modify, search similar entities and
define the changes to be proposed.
Definition2: An ontology change operation is a
set of change operation assigned to the ontological
entity OC ={oc
1,
oc
2
,oc
3,
....oc
n-1,
oc
n
}.
When the user adds a document, the ontology agent
detects the ontologies to modify. The ontology is
founded by similarity measure between the name of
ontology and the topic of document. If ontology
exists the ontology agent seeks similarities between
extracted terms and the ontological entity.
Depending to the level of similarity, the ontology
agent proposes the changes based on its knowledge
base (Chamekh, et al, 2013).
Example 3. Let’s take again the change made by
user 1. The ontology agent1 detects the ontology to
modify and seeks similarities between extracted
terms and the ontological entities of the ontology O.
In this case, it uses its knowledge base and proposes
that the concept Non infection inflammatory
pathology can be a sub concept to the concept
inflammatory pathology.
The management of the ontology consistency
and the conflict resolution are ensured by the
OntologyModificationinaMulti-UserConcept-ConflictResolution
299
consistency agent and the version agent. A detailed
presentation is given in section5.
4.4 Consistency Agent
The identification of types of change to apply on the
ontology, expresses formally the needs of evolution
required by users. When they are applied, the
ontology changes from a current version to another
one. However, the application of a type of change
could generate inconsistencies on the new ontology
version. In our framework, we assign the task of
keeping consistency to the consistency agent.
Two types of OWL consistency are
distinguished: structural consistency and logical
consistency. Structural consistency refers to
syntactic conditions of OWL DL constructors and to
the constraints specified on its elementary axioms
and their combinations (Djedidi el al., 2010).
Logical consistency refers to the formal semantic of
the ontology and thus, to its satisfiability in the
meaning that ontology is semantically correct and
does not present any logical contradiction.
Definition 3: we define consistency Q as a set of
consistency condition: Q= {q
1,
q
2
,q
3,
.....q
n-1,
q
n
}. An
Ontology O is consistent if it satisfies all the
consistency conditions Q(O).
Many reasoners are available to verify logical
inconsistency. However, some of them like Pellet
detects inconsistency but does not precise the axiom
that cause inconsistencies neither how to resolve the
detected inconsistencies.
In our work, we propose to detect and resolve
inconsistencies before they have occurred. So, in this
case the reasoner cannot be used.
To guarantee a logical and structural consistency,
we identify a set of rules. They are implemented as a
knowledge base of consistency agent (Chamekh, et
al, 2013).
Definition4: we define an additional change as a
set of operation that can keep the consistency of
ontology, AOC={ aoc
1,
aoc
2
,.....aoc
n
}.
The role of the consistency agent is keeping the
consistency of ontology during the process. Each
consistency agent is connected to another ontology
agent and another version agent in accordance with
the negotiation protocol. The negotiation allows the
consistency agent to perceive its environment. The
consistency agent has to achieve three objectives:
To denote a type of change by
perceiving the message sent by the ontology agent.
To study the effect of changes and
propose additional changes. The agent uses
consistency rules encapsulated as a knowledge base.
To propose the additional changes to the
user by communication with the user agent.
The general behaviour of consistency agent is
presented through the algorithm1.
The consistency agent perceives the proposed
change and the ontological entity from the ontology
agent. This one checks the consistency of ontology:
if inconsistency exists, the consistency agent
proposes additional change (based on its knowledge
base) to the user, if he validates the consistency
agent launches the task of creation new version;
otherwise, the change is cancelled.
Algorithm 1: general behaviour of consistency agent.
Begin
Perceive ()
OC={ oc
1,
oc
2
,oc
3
};
O ontology
C= {c
1
,c
2
,c
3
,....c
n-1
,c
n
}
Process verification of consistency ();
If (not Q(O) )
Process of denotation of consistency problem ();
If (aoc
1
€ AOC exist)
Process proposes additional change ();
Process to the user ();
If user validate
Process creation new version ();
Else
Undo change ();
End if
Else
Process to the user ();
End if
Stop agent ();
End if
End
Example 4: Let’s consider again a required
change presented up.The inconsistency agent1
studies the impact of this change (the concept Non
infection inflammatory pathology can be a sub
concept to inflammatory pathology) to the ontology
consistency by using the rules R1 and R2.
R1: if we add a sub-concept A to a non-leaf
concept B, we must compare the property of concept
A with the property of concept C (which is the sub-
concept of concept B). If there are common
properties between concept A and concept C, we
must verify R2, else (if not) we must verifyR3
R2: If there are common properties between the
concept A and the concept C
R3: If there are not common properties between
the concept C and the concept A
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
300
4.5 Version Agent
In a multi-user context, users add simultaneously the
new document. Agents run the process of ontology
evolution until the creation of new version. There
are two types of conflict situations: The simple
conflict situation and the complex conflict situation.
Definition5: we define a conflict situation as a
set of simple and complex conflict.
C={c
simple
, c
complex
}
C
simple
={c
s1
,c
s2
,......c
sn
}
C
complex
={c
c1
,c
c2
,.....c
cn
}
The agent behaviour is illustrated by algorithm 2.
Firstly, the agent receives the change, the concerned
ontological entity and the version of ontology.
Secondly, it verifies the current version; it applies
the change and creates the new version, if it is the
same version. Else the version agent verifies in the
log of changes the modifications that bound to the
ontological entity. The version agent searches in the
local rules the concerning conflict situation. If it is a
simple conflict situation, it resolves the problem, and
creates the new version and if it is a complex
situation conflict, the version agent sends a message
to the inconsistency agent with a proposed change.
When the conflict situation doesn’t exist in the rules
base, the version agent requests the validation of the
user and updates its knowledge base.
The simple conflict situation that can be directly
solve by the version agent.
Simple conflict situation 1:
The user 1 links or adds the instance I to the concept
A in the ontology O version X and in the log of
changes the property Y of the concept A has been
renamed in the property Y’. The version agent adds
or links the instance I to the concept A and creates
the new version.
Simple Conflict situation 2:
The user 1 rename the property Y (property Y’) to
the concept A in the ontology O version X and in the
log of changes the concept A has been renamed to
the concept A’. The version agent renames the
property Y (property Y’) to the concept A’ and
creates the new version.
Simple Conflict situation 3:
The user 1 links or adds the instance I to the concept
A in the ontology O version X and in the log of
changes the concept A has been renamed to the
concept A’. The version agent adds or links the
instance I to the concept A’ and creates the new
version.
Simple Conflict situation 4:
The user 1 adds the property Y or renames the
property Y (property Y’) to the concept A in the
ontology O version X and the concept A has been
renamed to the concept A’. The version agent
renames the property Y (property Y’) or adds the
property Y to the concept A’ and creates the new
version.
Simple Conflict situation 5:
The user 1 renames the property Y (property Y’) to
the concept A in the ontology O version X and in the
log of changes the property Y of the concept A has
been renamed to the property Z. The version agent
renames the property Z in to the property Y’ and
creates the new version.
Simple Conflict situation 6:
The user 1 adds upper-concept/sub-concept B to
concept A in the ontology O version X and in the log
of changes the concept A has been renamed to the
concept A’. The version agent adds upper-
concept/sub-concept B to the concept A’ and creates
the new version.
Simple Conflict situation 7:
The user 1 adds a synonym concept B to concept A
in the ontology O version X and in the log of
changes the property Y of the concept A has been
renamed to the property Y’. The version agent adds
a synonym concept B to the concept A and creates
the new version.
The complex conflict situation cannot be resolved
by the version agent. It sends proposed changes to
the inconsistency agent. This one uses the
consistency rules to solve the problem.
Complex Conflict situation 1:
The user 1 links or adds the instance I to the concept
A in the ontology O version X and in the log of
changes new instances have been added to the
concept A. The version agent sends a change
<add/link the instance I to the concept A> to the
inconsistency agent.
Complex Conflict situation 2:
The user 1 links or adds the instance I to the concept
A in the ontology O version X and in the log of
changes the instance I has been linked to the concept
B. The version agent sends change <add/link the
instance I to the concept A> to the inconsistency
agent.
Complex Conflict situation 3:
The user 1 adds upper-concept/sub-concept B to
concept A in the ontology O version X and in the log
of changes the sub-concept C or upper-concept D
has been added to the concept A. The version agent
OntologyModificationinaMulti-UserConcept-ConflictResolution
301
sends a change <adds upper-concept/sub-concept B
to concept A > to the inconsistency agent.
Complex Conflict situation 4:
The user 1 adds upper-concept/sub-concept B to
concept A in the ontology O version X and in log of
changes the property Y has been added to the
concept A. The version agent sends a change<adds
upper-concept/sub-concept B to concept A >to the
inconsistency agent.
Complex Conflict situation 5:
The user 1 adds upper-concept/sub-concept B to
concept A in the ontology O version X and in the log
of changes, the property Y has been renamed to the
property Y’. The version agent sends a change<add
upper-concept/sub-concept B to concept A >to the
inconsistency agent.
Algorithm 2: general behaviour of version agent
Begin
Perceive ()
OC= { oc
1,
oc
2
,oc
3
};
Vn ontology version;
Process verification version ();
If (Vn=current version)
Process create new version ();
Else
Process check change ();
If (OC not exist in log of change)
Process create new version ();
Else
Process search conflict situation ();
If (C
simple
)
Process create new version ();
Else
Process sends change ();
End if
End if
End if
Stop change ();
End.
Example 5.To illustrate a possible instantiation
of this conflict situation and let’s consider again a
required change Ch1 made by user 1 defining non
infection inflammatory pathology as a sub concept
of inflammatory pathology. After verifying the
consistency of ontology behind the change made by
user 1, the consistency agent sends the change, the
corresponding entity and the version of ontology.
The version agent verifies if it is the current version.
The current version is not the same version sent by
the consistency agent. The version agent verifies in
the log of changes if the entity inflammatory
pathology has been changed. It detects that this
entity has been renamed by another user to
inflammatory pathologies. The version agent detects
a conflict situation. It found that it can use the
simple conflict situation 6. The version agent creates
the new version.
5 CONCLUSIONS
In this work, we treat ontology evolution from
concepts extracted from documents. The different
steps of evolution phases are decomposed inside
different agents. We try to deal with the problem of
conflict situations in a multi-user context. So the
modifications on concepts (and ontologies) may be
performed in a parallel way.
Some conflict situations can be directly solved
by agents. Some of them must be sent to the
concerned agent who is in charge of modifications.
The first perspective is to formalize the different
conflict rules. Secondly, to define the process of
decision of each agent and finally, to confirm our
assumptions by the implementation of the system.
REFERENCES
Chen,Y, Peng, X and Zhao, W., 2009 An Approach to
Detect Collaborative Conflicts forOntology
Development, In the procedeegins of Advances in
Data and Web Management, Joint International
Conferences, APWeb/WAIM, China, pp.242-254.
Chamekh, F, Talens, G, Boulanger, D, 2013. Corporate
semantic web evolution: an approach based on multi-
agent system. In the procedeegins of International
conference on Management of computaionnal and
collective intelligence in digital ecosystems,
Luxembourg, pp. 235-240.
Chamekh, F, Talens, G, Boulanger, D, 2013. Ontology
evolution in the corporate semantic web. In the
procedeegins of International conference on
Entreprise Information Systems, Angers, France, pp.
182-189.
Djedidi,R, Aufaure, M.A., 2010.Onto-Evo an Ontology
Evolution Approach Guided by Pattern Modelling and
Qualit Evaluation, In the procedeegins 6th
International Symposium on Foundations of
Information and Knowledge Systems , LNCS: Vol.
5956, G.Berlin, Germany: Springer. Sofia, Bulgaria,
pp. 268-305.
Gandon, F., 2002. Distributed Artificial Intelligence and
Knowledge Management: Ontologies and multi-agent
systems for a corporate semantic web, PhD Thesis
INRIA, November .
Horrocks, I, Patel-Schneider,P.F and van Harmelen, F.
From SHIQ and RDF to OWL: The Making of a Web
Ontology Language .Journal of Web Semantics, 1(1),
2003.
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
302
Horrocks, I, Patel-Schneider,P.F and van Harmelen, F.
Reducing OWL Entailment to Description
LogicSatisfiability. Journal of Web Semantics, 1(4),
2004
Klein, M, 2004. Change Management for Distributed
Ontologies,PhD thesis,University of Amsterdam.
Luong,P-H, Dieng-Kuntz, Boucher, A., 2006. Managing
semantic annotations evolution in the CoSWEM
system, In Proceedings of the Third National
Symposium on Research, Development and
Application of Information and Communication
Technology, Hanoi ,pp. 215 –223.
Maedche,A . Staab,S., 2000. The text-to-onto ontology
learning environment, In Software Demonstration at
ICCS-2000-Eight International Conference on
Conceptual Structures.
Noy,N.F, Chugh,A, Liu,W, Musen ,M A., 2006. A
framework for ontology evolution in collaborative
environments, In Proceedings of the 5th International
Semantic Web Conference ISWC’06.
Rogozan,D ,Paquette, D., 2005.Managing ontology
changes in the semantic web, In the IEEE/WIC/ACM
International Conference on Web Intelligence,
Compiegne, France.
Rahman,S.D, Yadav, D, Agarwal, P, Bisht, P.S., 2012.
Multiagent Knowledge Management Architecture.
Journal of Software Engineering and Application ,
Volume 5,pp. 33–40.
Stojanovic,L.,2004., Methods and Tools for Ontology
Evolution, PhD thesis, University Karlsruhe.
Sellami, Z, Camps, V., 2012. Evaluation of a multi-agent
system for the evolving of domain ontologies from
text, In proceedings of International Conference on
Practical Applications of Agents and Multiagent
Systems (PAAMS ), Salamanca , p. 169-179.
Jaziri,W, Sassi, N, Gargouri, F., 2010. Approach and tool
to evolve ontology and maintain its coherence,
Metadata, Semantic and Ontologies, vol. 5, n°2, pp.
151-166.
Zablith, F, Antoniou,G, d’Aquin, M, Flouris,G,
Kondylakis,H, Motta,E , Plexousakis,D and Sabou,
M.,2014 Ontology Evolution: A Process Centric
Survey, In: The Knowledge Engineering Review,
Vol0:00, pp.1-31.
OntologyModificationinaMulti-UserConcept-ConflictResolution
303