Agent-SSSN: a strategic scanning system network based on multiagent intelligent system and ontology

Yman CHEMLAL Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 5, Issue 12, (Part - 3) December 2015, pp.143-150
www.ijera.com 143|P a g e
Agent-SSSN: a strategic scanning system network based on multi-
agent intelligent system and ontology
Yman CHEMLAL *, Hicham MEDROMI *
*(EAS- LISER) Systems Architecture Team, ENSEM) - Hassan II University, Casablanca, Morocco
ABSTRACT
This article reports a development of a strategic scanning system network prototype system based on multi agent
system and ontology, called Agent-SSSN, for developing business intelligent strategies. This is a cooperative
approach to integrate the knowledge of experts in business intelligent system. The approach presented in this
chapter is targeted towards using ontologies. The use of ontologies in MAS environment enables agent to share
a common set of concept about context, expert user profiles and other domain elements while interacting with
each other. In this paper, we focus especially on the modeling of the system Multi-Agents using O-MaSE
(Organization-based Multiagent Systems Engineering Methodology) and a conceptual diagram of the ontology
database.
Keywords : business intelligent, multi-agent intelligent system, ontology, O-MaSE, strategic scanning system
network.
I. INTRODUCTION
There is a growing recognition in the business
community about the importance of knowledge as a
critical resource for enterprises. The purpose of
knowledge management is to help enterprises create,
derive, share and use knowledge more effectively to
achieve better decisions, increase of competitiveness
and fewer errors. In order to run business effectively
an enterprise needs more and more information
about competitors, partners, customers, and also
employees as well as information about market
conditions, future trends, government policies and
much more.[2]
Different applications within information
systems (IS) that support wide range of
functionalities need to be integrated in order to
provide the appropriate level of information support
[6]. One of the prominent approaches for IS
integration is the use of Business Intelligence (BI)
processes and Multi-Agent Systems (MAS).
Business intelligence applications are not a new
trend any more, but they have become a must during
the last decade as a basic tool used by the modern
management. Business intelligence is the result of
the natural evolution in time of decision support
systems and expert systems, systems that aimed at
replacing humans in the decision making process or,
at least, at offering solutions to the issues they are
concerned of. Gartner’s definition
Integrating Business Intelligence (BI) processes
in an information system requires a form of strategic
scanning system for which the information is the
main source of efficiency and decision support [1].
The strategic scanning system is based on a
strong idea: any actor in the company is likely to
hold information elements and it is the synergy of
these given elements which rise to usable
information for action [7].
“Business Intelligence as a strategic scanning
system network is the collective process by which a
group of individuals track down and use anticipatory
information about changes likely to occur in the
external environment of the company, in order to
create business opportunities and reduce risk and
uncertainty generally” [11].
So according to the definition a strategic scanning
system network is the network of observers that
are attached to the group of expert and provide it
informal information collected for example from
customers, suppliers, competitors [4].
It is believed that, for the business intelligence (BI)
of an enterprise, only about 20% of information can
be extracted from formatted data stored in relational
database. The remaining 80% of information is
hidden in unstructured or semi-structured documents
[5]. But relevant information gathering in the web is
a very complex task. The main problem with most
information retrieval approaches is neglecting the
context of the pages [14]. For that reason in
Business Intelligence we introduce a specific
cooperative information gathering approach based
on the use of software agents and ontologies to
represent the knowledge of experts, named
AGENT- SSSN.
RESEARCH ARTICLE OPEN ACCESS

Multi Agent technology is often mentioned as an
approach to design and develop flexible and
distributed software systems [3]. The development
process of a Multi System Agents consists of four
stages namely the analysis, design, development and
deployment [9]. Many methodologies allowing the
development of SMA, have been proposed. Among
which, we chose the O-MaSe methodology to
develop our SMA. This methodology is selected for
its simplicity.
The use of ontologies in MAS enables agents to
share a common set of facts used in user profiles,
products and other domain elements, while
interacting with each other. With exploiting
reasoning mechanisms new knowledge can be
derived from known facts and improve the
Knowledge base (KB).
Organization of this paper is as follows: Firstly, the
presentation of the problem studied by giving an
overview of the state of the art. Next, in section 3,
the multi-agent architecture of the Agent-SSSN
system is presented with the roles of agents and
ontologies. Finally, the conclusions are drawn with
further research work envisaged.
1. Related work:
The Business Intelligence aims at enabling decision
makers and managers of a company to have
strategic Information. It therefore involves learning,
observing, understanding an external environment
and apprehending the events to make the best
decisions for more competition and innovation. This
involves a several approach deal with agent and
ontology to improve a process of business
intelligence (BI).
Research in (kishore, zhang.2006) [19] showed that
MAS provides an excellent approach for modeling
and integrated business IS. In (Soo, Line.2006) [20]
authors propose a cooperative MAS platform allows
the invention process to be carried out though the
cooperation and coordination among software
agents delegated by the various domain experts in
the complex industrial R&D environment. In
(Olivier Chator, Jean-Marc Salotti. 2012) [21], a
solution focused on the implementation of a
collaborative tool based on MAS where agents are
skills, has been proposed. Authors in (Espinasse,
Fournier, Freitas .2006) [14] conducted an
interesting research about collecting and extracting
relevant information gathering in the web based on
software agents and ontologies. In (Amos David,
Sahbi Sidhom. 2005) [15], a proposal focused on
some models and tools for the implementation of the
complex processes of Business Intelligent, has also
been used. In (Dejan Lavbic. 2012) [5] author
propose an approach in integration of unstructured
information found in the web with information
available in several internal data sources. In
(Maizura, Rosmayati. 2010) [17] architecture for
MAS based decision Support Systems has been
used.
2. Proposal for solution
The review of related work presented in previous
section focused on improving how multiagent
systems and software agents have been applied in BI
and how can their potential be used for business
intelligence systems improvement or on domain of
research and collect information in the web
independently on the BI, However, these researches
do not consider the process of strategic scanning
system in the Business Intelligence to realize the
routine monitoring of technological sectors and the
market competition then the rational use of captured
data.
This chapter presents a proactive process of
observation and analysis of the environment for
integration a strategic information in Enterprises,
using agent oriented approach based on ontologies.
Fig1
This approach focuses on a collection and extraction
of information domain specific from the Web based
on software agents, with a consideration of the
context of the search using the ontology and the
information dissemination by simply delivering the
right information at the right time to the right expert
in the system.
Fig.1 Global System Architecture

2.1 The role of ontology
Ontology represents the domain knowledge and can
be used to support various processes within a multi-
agent system. Ontology is high expressive
knowledge models and as such increase the
expressiveness and intelligence of a system.
Ontology provides a common way of understanding
between different agents.[18]
Nowadays Web-based information retrieval systems
are widely distributed and deeply analyzed from
different points of view. The main objective of all of
such systems is to help users to retrieve information
they really need (obviously as quickly as it is
possible) [16], then the combination of the multi-
agent approach with declarative knowledge, leading
to the use of ontologies is relevant for the
development of Web-based information retrieval
systems.
The knowledge of the expert is crucial to the
interpretation of extracted patterns, one of the
challenges of Agent-SSSN system is to collect and
extract information that is interesting and useful for
expert users.
We used in our research the knowledge management
approach where every agent has knowledge about its
own research context, our system is based on the
semantic representation of the knowledge used by
agents:
According to [6] ontology can be structured into
different sub-ontologies –upper ontology, domain
ontology, task ontology and the application
ontology. Following similar guidelines we have
defined upper ontology named organization
domain and combined domain and task ontologies
in monitoring ontology, notify ontology and
specific ontology, Organization ontology is limited
to abstract concepts and it covers reusable
dimensions. Task ontologies specify concepts of
monitoring ontology, notify ontology and
application ontology specify a specific ontology.
(see Fig. 2).
Fig 2. Modularization of Ontologies depending on the
scope and partial ordering defined by inheritance
Agent-SSSN integrates expert knowledge
throughout the process of information retrieval on
the web; knowledge is structured and organized in
the Knowledge Base (KB). Our system is based on
the semantic representation of expert knowledge,
For this we use the proposed clustering of
ontologies is based on the common understanding of
the business activity domain being defined in
organization ontology. Every agent has its own
interpretation of a Knowledge Base (KB), which is a
specialization of Organization ontology with detail
definition of knowledge required by individual
agent.
2.2 Knowledge modeling
Ontologies are not available; they must be
constructed specifically by asking experts. Their
role is to provide a common vocabulary to several
agents and to facilitate a particular context, in this
case the search and retrieval of information on the
Web based on the expert knowledge.
The construction of ontology goes through the
following phases:
- Delimit the domain of interest and the level of
abstraction to describe it,
- Define specific vocabulary knowledge of the
domain, that is to say a set of terms of assertions and
semantic constraints on the domain,
- Model the knowledge in terms of concepts and
taxonomy of individuals, relationships between
them, constraints and inference rules.
Fig 3 Modularization of Ontologies
2.3 Construction of the knowledge base
In the present version of our prototype, all
ontologies are combined in a single knowledge base
.This part of the system include the information
shared between agents and interpreted by the group
of experts.

It is composed of three main families information:
organization domain : subsumes all concepts of
global ontology concepts for the activity domain of
the organization
monitoring ontology: groups all the expert
knowledge about monitoring activities (sales,
marketing, competitive…)
notify ontology: All knowledge about notification
is defined in notifying ontology, where every expert
user has his own context defined and the position
within organisation across two dimensions –
organisational unit (e.g. Marketing, Sales, Human
resources etc.) and decision making level.
specific ontology: includes all concepts is the
domain of the organization's activity and linked to a
specific monitoring domain.
Web ontology: define web concepts.
The main concepts of the knowledge base, shown in
Fig 4, using the UML formalism are: Concept-OD,
Concept- MO, Concept-NO, Concept- SO.
Each cluster is attached to a single domain defined
by a specific ontology.
Fig 4. Main classes and relations of the ontology
2.4 The role of agent
Strategic scanning system network or Network
knowledge production is a complex system. This
complexity is particularly due to the collect public
data that are freely available on Web in order to
support the decision-making process. Information
agent is capable of collecting information and
organizes it as a local database.
The multiagent approach offer modular, flexible,
scalable and generalisable algorithms and systems
solutions for information retrieval, extraction,
fusion, and data-driven knowledge discovery using
heterogeneous, distributed data and knowledge
sources in information rich, open environments [17]
.Indeed, this approach relies on the assumption that
a system is composed of autonomous entities, called
agents, that interact in order to deal with a global
goal or some local tasks. Intelligent software agents,
as a new artificial intelligence technology, may
bring benefits to the strategic scanning system
network [13]. In this study, the Agent-SSSN system
is based upon the task-sharing framework and the
task-sharing are assigned to relevant strategy agents
and using ontologies for several tasks, that ontology
is used by every agent to represent the interpretation
of a domain observer or relevant information for
network observers but also for communication
between agents and the network of experts.
To model our system Multi-Agents, we used the
methodology O-MaSE [Deloach, 2005]. O-MaSE
(Organization based Multi-Agent System
Engineering) is a extension methodology MASE
[Deloach 1999] which completes the organizational
dimension. O-MaSE considers a multi-agent system
as an organization social. Each agent is a member of
this organization and plays a specific role according
to its capacity. It is composed mainly models (goal,
Organization, Roles, Ontology, Agent, Agent
Protocol and State). [9]
2.4.1 Goals Diagram
In our problem, the main goal is to integrate an
observing network for collecting information on
restricted domain of the web. This objective is the
global goal “goal 0”. The “goal 0” dependent on the
completion of another two sub-goals which are:
cooperative information gathering “goal 1”,
Broadcast information “goal 2”. “goal 1” depends
in his turn on the implementation of many authors
goals. Indeed, to collect information, we must start
by search information “goal 1.1”, analyze
information “goal 1.2”, extract information “goal
1.3” and storage information “goal 1.4”.
Fig 5. Goals Diagram
2.4.2 The multi-agent architecture of the
Agent-SSSN
The starting point of this architecture is a prototype
already achieved the system AGATHE. This system
already adopted the agent and also uses ontologies

approach to realize classification and information
extraction tasks on the Web using a specific
ontology, but as previously mentioned our approach
develop a conceptual ontology stored in knowledge
base (KB) and instanced by a specific agent
described in agent platform table. See table 1
2.4.3 Agent’s platform
Our objective is to describe the different roles that
can be exercised by the agents of the system. For
each goal /sub goal previously identified, we must
create a role for achieving this.
In order to achieve a goal, a role must have at his
disposal one (or more) ' capabilities' which translate,
usually by execution plans (Diagram states /
transitions describing the manner in which an agent
should behave).
As described previously, five goals consist our goals
diagram. To realize them we identified five roles.
Each class represents a model for a type of agent
that can be instantiated many times according to the
system requirements.
Goals Roles Capabilities Agent Ontology
Goal 0: observing
network for
collecting
information on
restricted domain
of the web
management of
dynamic
interactions
between
network actors
Activate the search
process by creating a
Information Retrieval
Agent,
supervise the flow of
information, efficient
running of the
negotiations to
synthesize and analyze
information, it is capable
according to the needs
of creating agents or
delete,
Mediator agent Organization
ontology
Goal 1.1 search
information
Play
information
retrieval role
for the purpose
of decision
making.
querying external search
engines on the Web (like
Google) to obtain web
pages
Information
Retrieval Agent
Specific
ontology
Goal 1.2: analyze
information
Syntactic
analysis
analyze and classify the
web pages according to
a specific ontology
Analyst agent Web
ontology
Goal 1.3: extract
information
Classify and
extract
information
A semantic
classification of pages
received, as well as
information extraction.
Each extractors agents is
associated with a
particular concept in the
specific ontology
extractor agent Specific
ontology
Goal 1.4: Storage
information
Storage
information
processes the
information received in
order to comply with the
Database agent Specific
ontology

format suitable storage
device according to the
database structure
Goal 2: Broadcast
information
delivering of
the right
information at
the right time
and the right
expert user
Information is
proactively delivered by
agents to the expert
without a specific
request. All knowledge
about notification is
defined in Notifying
ontology, where every
expert user has his own
context defined and the
position within
organization across two
dimensions.
in our approach we
implemented the “push
model”
Notification
Agent/ Push
agent
Notify
ontology
Table 1. Agent platform table
The whole architecture is depicted in Figure 6.
Fig 6. The global architecture
2.5 The logical flow of the Agent-SSSN
system execution
The logical flow of the Agent-SSSN system
execution is shown in Fig. 7 which is briefly
described below:
Step1: a Mediator agent activates cluster of
Information Retrieval Agent, Analyst agent,
extractor agent and assigns to him a concept of a
specific ontology for a specific cluster.
Step 2: a group of Information Retrieval Agent
(IR) transfer the request to the various search
engines and collects the resulting WebPages
Step 3: These WebPages are transmitted to the
Analyst agent for treatment.
Step4: Analyst agent makes a syntactic analysis
the web pages according to web ontology.
Step 5: Extractor agent receives pages from
Analyst agent, makes a semantic classification
according to a specific ontology. Each Extractor
agent is associated with a particular concept in the
specific ontology associated with specific cluster.
Step 6: the database agent processes the
information received from Extractor agent in order
to comply with the appropriate format storage the
database structure.
Step 7: According to the expert user profile (using
notify ontology), push agent sent a notification by
using several technologies from Windows Alert, e-
mail, Really Simple Syndication (RSS), Short
Message Service (SMS) etc. These notification

types are also ordered by priority for each expert
user and according to this type the content is also
adapted.
Fig 7. logical flow of the Agent-SSSN system
Conclusion & future work
This study has been sought to explore how the
process of strategic scanning system network
development
can be improved by multi-agent intelligent system
based on ontology. To achieve this aim a platform
able to integrate a process of strategic scanning
system network in the business intelligence (BI)
called Agent-SSSN prototype system has been
created and described. This approach focuses on a
collection and extraction of information that is
interesting and useful for expert users. Sources of
information discussed in this approach are the
information gathering in the web.
As future enhancements we would like to:
- extend our model by defining a detailed
architecture of each agent, specifying
communication between these agents using ACL
(Agent Communication Language) and
implementing the architecture proposed in the
SMA development platform Jade (Java Agent
Development Framework), selecting OWL (Web
Ontology Language) as a language for ontology
presentation.
- Evaluate the efficiency and effectiveness of this
proposal by implementation of our approach as
architecture from a real case study.
References
[1] Y.CHEMLAL, H.MEDROMI, Improving
the quality of information in strategic
scanning system network: approach based on
cooperative multi-agent system,
International Journal of Artificial
Intelligence & Applications (IJAIA), Vol. 6,
No. 1, January 2015.
[2] Fuentes, V., J. Carbo, Heterogeneous
domain ontology for location based
information system in a multi-agent
framework, Intelligent Data Engineering
and Automated Learning - Ideal 2006,
Proceedings Vol. 4224, No., pp. 1199-1206.
[3] Alexander Patrick Joseph Loebbert , Multi
Agent Enhanced Business Intelligence For
Localized Automatic Pricing In Grocery
Chains, MIT(Honours), MBA, Int. Dipl.
Betriebswirt, decembre 2011.
[4] l’intelligence économique.. La comprendre,
L’implanter, L’utiliser, Deuxième tirage
2006, © Groupe Eyrolles, 2006, pour la
nouvelle présentation, ISBN : 2-7081-3604-
6.
[5] Dejan Lavbič, Knowledge Management with
Multi-Agent System in BI Systems
Integration, E-Business - Applications and
Global Acceptance, ISBN 978-953-51-0081-
2 Hard cover, 136 pages, Publisher InTech,
Published online 10, February, 2012,
Published in print edition February, 2012.
[6] Guarino, Formal Ontology and Information
Systems. FOIS'98, Trento, Italy, IOS Press.
N. (1998).
[7] la veille stratégique du concept au pratique,
Institut Atlantique d’Aménagement des
Territoires (IAAT) (2005).
[8] LINK-PEZET Jo, R.BERKANE, D.
MOTTAY, intelligence économique et
décision.

[9] Fethi Mguis, Kamel Zidi, Khaled Ghedira,
Pierre Borne, Modélisation d'un Système
Multi-Agent pour la résolution d'un
Problème de Tournées de Véhicule dans une
situation d'urgence, 9th International
Conference on Modeling, Optimization &
SIMulation, Jun 2012, Bordeaux, France. 7
p.
[10] Report "CGP IE" (General Commissariat of
Intelligence Plan), named in his use name
"Marten Report" and published in the Report
of the Commissioner General Plan in 1994.
[11] Veille stratégique, Concepts et démarche de
mise en place dans l’entreprise, Humbert
LESCA.
[12] Approche dynamique de l'intelligence
économique en entreprise : apports d'un
modèle psychologique des compétences :
Contribution à l`élaboration de
programmes d`actions de la CCI de Rennes,
Psychologie. Université Rennes 2;
Université Européenne de Bretagne, 2010.
[13] Shuliang Li, AgentStra: an Internet-based
multi-agent intelligent system for strategic
decision-making , 2006 Elsevier Ltd.
[14] B. Espinasse, S. Fournier et F. Freitas,
aghathe : une architecture générique à base
d’agents et d’ontologies pour la collecte
d’information sur domaines restreints du
Web
[15] Amos David, La recherche collaborative
d'information dans un contexte d'Intelligence
Economique.
[16] Garces, P. J., J. A. Olivas, et al. (2006).
Concept-matching IR systems versus word-
matching information retrieval systems:
Considering fuzzy interrelations for indexing
Web pages, Journal of the American Society
for Information Science and Technology
Vol. 57, No. 4, pp. 564-576.
[17] Noor Maizura Mohamad Noor and
Rosmayati Mohemad (2010). New
Architecture for Intelligent Multi-Agents
Paradigm in Decision Support System,
Decision Support Systems, Chiang S. Jao
(Ed.), ISBN: 978-953-7619- 64-0
[18] Dr. Prashant M. Dolia ,Integrating
Ontologies into Multi-Agent Systems
Engineering (MaSE) for University
Teaching Environment, journal of emerging
technologies in web intelligence, vol. 2, no.
1, february 2010
[19] Kishore, R., H. Zhang, Enterprise integration
using the agent paradigm: foundations of
multi-agent-based integrative business
information systems, Decision Support
Systems Vol. 42, No. 1, pp. 48-78.2006
[20] Soo, V. W., S. Y. Lin, A cooperative multi-
agent platform for invention based on patent
document analysis and ontology, Expert
Systems with Applications Vol. 31, No. 4,
pp. 766-775.2006
[21] Olivier Chator Jean-Marc Salotti,
Modélisation Multi-Agents centrée sur les
Compétences pour la Collaboration des
Acteurs dans les Projets de Développement
Durable.2012
Authors
Yman CHEMLAL is an engineer in computer
science from the INPT, in July 2005, Rabat,
Morocco. She prepares her thesis at ENSEM,
Hassan II University, Casablanca, Morocco.
Hicham Medromi received the PhD in engineering
science from the Sophia Antipolis University in
1996, Nice, France. He is responsible of the system
architecture team of the ENSEM Hassan II
University, Casablanca, Morocco. His actual main
research interest concern Control Architecture,
Architecture of
System and Software Architecture Based on Multi
Agents Systems and Distributed Systems. Since
2003 he is a full professor for Control systems and
computer sciences at the ENSEM, Hassan II
University,
Casablanca. He managed eight Research projects
and he has published several Patents and
publications in international journals and
conferences.

Agent-SSSN: a strategic scanning system network based on multiagent intelligent system and ontology

Recommended

More Related Content

What's hot (18)

Viewers also liked (20)

Similar to Agent-SSSN: a strategic scanning system network based on multiagent intelligent system and ontology (20)

Recently uploaded (20)

Agent-SSSN: a strategic scanning system network based on multiagent intelligent system and ontology