JADE AGENT IS INTERMEDIATION SYSTEM (JAIS) FOR

KNOWLEDGE EMERGENCE IN COMMUNITY OF PRACTICES

Intelligent Information Systems, Best Practices & Communities of Practice,

Cooperation, Communication, Collaboration and Knowledge Sharing

Clauvice Kenfack

MODEME, Jean Moulin University,6 cours Albert Thomas, Lyon, France

Keywords: Community of practices, Knowledge emergence, Intermediation system, JAIS agent.

Abstract: This paper presents an intermediation agents system to manage the distributed collaborative design

environment like CoPs. The JADE Intermediation System (JAIS) uses community enactment mechanism

and agent integration mechanism. The community enactment mechanism is the system kernel and follows

the specifications of the CoPs reference model. The system kernel supports two types of agents (human

agent and artificial agents) that help to manage the activity into the community, whereas the integration

mechanism supports an intermediation agent to interact, coordinate and monitor the activities between

agents. JAIS facilitates the team interaction in a collaborative and distributed environment.

1 INTRODUCTION

We describe an intermediation system able to design

distributed and collaborative environment in a CoPs.

JADE (2006) is used as the agent platform for

linking the heterogeneous system in a distributed

environment like CoPs. Gaia methodology is used in

our work to describe a part of MAS model and

MAS-CommonKADS to represent the knowledge

model. Agent is able to acting autonomously,

cooperatively, and collectively. Hammond

(Hammond and al, 2004) proposed an approach

agent to model knowledge virtual communities

implanted on JADE. The information increasing

across businesses involves a need to exchange and

share information from various distributed and

heterogeneous sources, (Boulanger, Dubois,). These

exchanges include the concepts of knowledge

acquisition, sharing and emergence, and require the

development of negotiation and interaction

protocols. According to Wenger (Wenger, 1998),

CoPs are "resources in the most versatile and

dynamic enterprises and (CoPs) form the basis of

cognitive ability and learning organizations". Some

authors such as Brown and Duguid (Brown and

Duguid, 1998) see in these communities a place for

knowledge creation, maintenance and reproduction.

In this context, we propose to design a system

composed of several layers: the first contains the

description of the organization (organizational

model), the second layer represents a

"agentification" of the previous level with a

modeling agent. The last level is the level of

implementation, will be processed using the JADE

platform. Before presenting our work, we give the

definition of the used concepts.

2 COPS AND KNOWLEDGE

EMERGENCE

This concept is often used in order to create

knowledge (experience, knowledge internalized,

tacit knowledge (Nonaka, 1991)), (Cohendet et al,

2003). Zacklad (Zacklad, 2003) argued that

knowledge is still a large part tacit and

contextualized. So we define CoPs as "a creative

community of knowledge formed a group of human

agents with a common interest in a given subject and

exchanging knowledge in connection with his

problem" (Kenfack, 2007). CoPs can be treated as an

intelligent system in which players do not work

alone but in an environment that contains other

intelligent entities. Therefore agent technology is a

good candidate to model CoPs because he offers a

372

Kenfack C..

JADE AGENT IS INTERMEDIATION SYSTEM (JAIS) FOR KNOWLEDGE EMERGENCE IN COMMUNITY OF PRACTICES - Intelligent Information

Systems, Best Practices & Communities of Practice, Cooperation, Communication, Collaboration and Knowledge Sharing.

DOI: 10.5220/0003103603720377

In Proceedings of the International Conference on Knowledge Management and Information Sharing (KMIS-2010), pages 372-377

ISBN: 978-989-8425-30-0

 2010 SCITEPRESS (Science and Technology Publications, Lda.)

great flexibility concerning development of complex

and distributed systems. An agent is a software

entity that can autonomously perform routine tasks

with a level of intelligence (Boudriga 2004)

Wooldridge (Wooldridge and Jennings, 1999).

During the problem solving, the capacity of all

individuals are used to act together. In the case of

CoPs, the collective knowledge available in the

community is much more than the sum of

knowledge of all members (Brown and Duguid,

1991). CoPs have a particular method of problem

solving; members of the community put existing

knowledge into new contexts and create new

knowledge. Therefore knowledge emerges through

communication and interaction between members of

the CoPs. The common enterprise emerges from a

permanent collective process during negotiation that

reflects the complexity of the dynamics of the

mutual commitment of community members. The

phenomena of knowledge creation emerge at the

macro level (the community as a whole) from an

interaction between the members of the CoPs (micro

level). Another aspect of the knowledge emergence

in CoPs can be realized through the techniques of

storytelling (Soulier, 2004).

Another concept that we use is the concept of

intermediation, we define our intermediation system

as "a system that allows all members of the CoPs to

create interactions between them, even in

geographically dispersed locations. To promote the

co-construction of meaning, to enrich their

knowledge base; to improve their skills, to share,

exchange and acquire knowledge. To do this, the

system must be able to provide the mechanisms to

manage constraints that could impose the

functioning of the community. These constraints can

be tasks allocation, management profiles,

authorization of access to the knowledge base,

process requests etc". (Kenfack, 2007). Our aim is to

reproduce some process to support human

intelligence by building intelligent intermediation

agents to perform members’ tasks in the CoPs. Our

proposal. This specification has identified and

defines the system models. The organizational

model defines knowledge and role models, the MAS

model defined communication and agent models. In

this figure we identified three phase of development

of our system.

Figure 1: JAIS Model.

3 JAIS PHASE

3.1 Identification Phase

In this phase we identified some components which

included some technical features to allow

information exchange, mutual support and mutual

point of view between the members of CoPs to

achieve and create the emergence of new

knowledge. These components are: component-

based resources used to save the formalized part of

the shared directory. It is important that the link

between the actor and the resource recorded or

published is maintained and easy to follow. The

system must encourage and store all information

produced by the community, whether a result of

interactions between peers, or individual

contributions.

The activities domain component: This

component allows the community to select its

members, organize the implementation of its

activities in a field, search for information resources

related to their activity domain.

The role component: This feature allows

members to define and identify the roles may be

held within the community or other communities. In

these roles a set of activities to achieve is assigned.

Each role includes objectives to be achieved by the

member who play this role. To these objectives

rights are assigned.

The basic Protocols / Collaborative Activities /

cooperatives Tasks: The protocol defines the steps

by which the activities must be performed by

community members, how the activity is done, what

kind of knowledge or skill is necessary to carry out

the activity. Cooperative activities represent

concrete actions that are undertaken in the

community. The authentication database: this

JADE AGENT IS INTERMEDIATION SYSTEM (JAIS) FOR KNOWLEDGE EMERGENCE IN COMMUNITY OF

PRACTICES - Intelligent Information Systems, Best Practices & Communities of Practice, Cooperation, Communication,

Collaboration and Knowledge Sharing

373

database contains the identifiers of each member in

connection with their profile and management rules

of each right.

User profile base component: This component

contains a list of all CoPs members that are

registered as users. Especially, in virtual CoPs face-

to-face contacts is not present, the design of profiles

user base allow building trust among members. We

propose to define the profiles in two areas: static

area, which include information background and

dynamic area which enable dynamic evolution their

profiles based on interventions in the community

Exchange component: This component allows

members to communicate between them and the

system through various channels (direct

asynchronous communication between members

who send messages directly to another members

they wish to contact or indirect asynchronous

communication between geographically dispersed

members.

Tool/collaborative platform: the collaborative

platforms like (SweetWiki, CommKnowledge) are

mostly equipped with one or more forums, more or

less formal in which members of the CoPs can

discuss various topics. Especially in this context

people can exchange experiences (through narratives

(storytelling) or by sending request for information

or assistance).

Analyse phase

The Organizational model

The organizational model represents the core

framework of a CoPs. The system specification

relies on the functioning of the community

especially on the relevant aspects of what constitutes

the community (activMember), the objectives,

policies put in place to support the community.

The knowledge model describes the knowledge

required for workers to perform activities. It also

helps to define the process to resolve activities

within the CoPs, to describe the activities and tasks

and their distributions, the inputs and outputs, the

preconditions and the performance criteria. So the

knowledge model is composed of:

The organizational structure: includes the

Organizational Context (background activity)

corresponds to the organizational environment in

which the community is involved; this allows

targeting the utility and function of the tool and its

suitability to the needs of users. The user profile: A

profile is define as including knowledge necessary

for effective evaluation of applications and

production of relevant information distributed to

each user. The goals are used to model the system

through the missions and goals of a community. The

members. Members can be defined as having

strategic and intentional goals. A member may be a

physical agent or a software agent.

The Activities domain: refers to topics related to

the environment in which the community is

immersed, this structure allows to identifying some

concepts.

The activities structure includes: The activity

context corresponds to the organizational

environment in which the community evolves; the

activity Protocol, the tasks, the cooperative

activities: a description of a cooperative activity is

independent of the entities that perform. The

cooperative tasks, represent the work being done by

members of CoPs, solving tasks contributes to

community resources.

Resources base: represent all library resources

and the knowledge exchanged and used the mails,

the group summaries and other information that may

help in solving a problem in the community. The

resource base is used to perform a cooperative

activity.

The role model identifies the key roles of the

system (Wooldridge et al. 2000). A role is defined as

an abstract entity which describes the functions in

the organizations; role describes an organizational

structure according to a type of activity (Gasser,

2001). In a role-model, role descriptions define or

identify activities or services required to achieve the

community goals. Indeed, a set of roles is attached to

each community and a member can play several

roles resulting in a set of actions necessary to

accomplish a task (the principle of agent-group-role)

(m association -> n) (Gutkchnet, 2001).

In the system, several types of roles can be

identified: an organizational role describes all the

roles that constitute the computational organization

(in a CoPs, and resulted in the intermediation

system) and an intermediation role. Each role

includes objectives to be achieved by the Member

who will play this role. The roles in our system are

characterized by two types of attributes according to

GAIA (Wooldridge et al, 2000) method: the

permissions represent the rights associated with the

role and on the type and number of resources to

exploit to achieve its objectives.

Responsibilities represent the goals, objectives of

a role defined by various features. The protocols are

activities which involve interactions with other

agents.

MAS model (Intermediation)

This model helps to reproduce the process able to

support human intelligence. The use of intelligent

agent helps to perform CoPs tasks. The following

KMIS 2010 - International Conference on Knowledge Management and Information Sharing

374

model covers all the models needed to develop our

intermediation system. The MAS model describes

the roles defined by the roles “agentification" in the

organizational model. Thereafter we describe

models of the MAS model.

The Role model in the system described

intermediation roles correspond to the

transformation of the organizational role from the

organizational model. These roles correspond to the

“agentification” activities at the organizational level.

In our system we count many roles : Dialogue

Manager Role manages the communication between

different agents system (via the technological tools

used in the CoPs). The Cooperative Activities

Manager Role: in charge of requests execution. They

manage the interactions during tasks solving. The

Knowledge base manager role supports management

resources of CoPs and agent, and this through:

Profile manager role; resources Manager Role:

protocol activities manager role; authentication

manager role.

The Communication Model: Defined the

activities between the actors (human and software

agent), their roles, and the resources they need to

perform their activities. This model allows

understanding and implementing communications

situations and describes communication links

between different agent types.

Jade follows FIPA standards so that, ideally,

Jade agents can interact with agents in other

languages and running on other platforms. There are

inevitably dependencies and relationships between

different roles in multi-agent system. It is therefore

necessary to represent these relationships, these

protocols in the communication model (Wooldridge

et al. 2000). This model is composed of messages,

protocols and interactions. Each role in the

interaction model becomes a communication link

between agents with similar objectives, the same

inputs and outputs. Interaction Protocol specifies the

interactions between agents. The interaction protocol

is specified in the structure of the interaction

activities of the system in terms of agreements

between the roles played by members during the

interactions. With this protocol, each activity

becomes an interaction process.

Agent model: This model defines the type of

agents in the system and structure of these agents

depending on the tasks entrusted to them. A type of

agent is derived from an agent or a set of roles. The

agent of this model ensures the control and

responsibility on the cooperation activities; this

model is composed of: the skills and knowledge.

Conceptual phase

This phase will define the technical components

of the intermediation model and implement the

interaction between members of the intermediation

model and of the JADE platform. In this model the

components are necessary to implement the goals of

the system will be represented. These are the aspects

concerning the organizational structure, business

cooperatives, agents, communications and

knowledge. We count five groups of agents: The

access manager agent is “agentified” and supports

two types of roles, the manager dialog and the

authentication role, the activity manager agent is

“agentified” and supports one type of role, business

manager role. The knowledge manager agent is

“agentified and support three types of roles, the

manager profile role, the resource manager role, the

protocol activities role. The transformation of the

CoPs model to an agent-based model begins with the

modeling of each member in the system. The

application of intermediation model will implement

the following technical features.

Example of application components to CoPeR

The agentification features will focus on the

components specified in the identification phase, we

show through this scheme at what level agents

intervene.

The agents we use have the capacity to handle

their tasks or solve problems. That’s means agents

have the ability to explicit knowledge representation

and communication (Huhn, 1999). The process of

"agentification" intermediation system is to identify

the characteristics of intermediation model from the

functional needs of the intermediation system.

The design process of multi-agents using the

JADE platform is to identify, plan and services of

each agent, behaviours and corresponding messages

exchanged between agents. Therefore, mechanisms

for communication between JADE agent

(Bellifemine, and al, 1999) systems used by agents

to communicate will be explored by identifying the

elements necessary for their implementation.

Cooperation between community members and the

whole system of intermediation is done through the

moderator an intermediation agent (agent manager

access). Two types of agents will coexist in our

system, there are also many human agents, including

ActivMember agent, the agent asked, and another

member later intermediation which is divided into

various categories of intermediation agents.

JADE AGENT IS INTERMEDIATION SYSTEM (JAIS) FOR KNOWLEDGE EMERGENCE IN COMMUNITY OF

PRACTICES - Intelligent Information Systems, Best Practices & Communities of Practice, Cooperation, Communication,

Collaboration and Knowledge Sharing

375

Figure 2: Example of cooperation between CoPeR Members.

4 CONCLUSIONS

In this article we presented a three-tier architecture

to extract, represent and the emergence of explicit

knowledge and tacit or transmitted in CoPs.

Modeling agent was privileged strate 2 and 3

because the agents have characteristics of flexibility,

learning and knowledge exchange through the

protocols of interaction and negotiation. Following

our work focuses on the complete implementation of

our proposal in particular the implementation of

interaction protocols. When communities become

large (several hundred), a backup centralized /

decentralized knowledge is desirable (in terms of

reuse).

Currently we are focusing on building a shared

ontology and a consensual own CoPs.

In this article we have presented a three-tier

architecture to extract, represent and the emergence

of explicit and tacit knowledge in CoPs. Modeling

agent had the privilege layers 2 and 3 because the

agents have characteristics of flexibility, learning

and knowledge exchange in the interaction

protocols. Our work focuses on the full

implementation of our proposal, in particular the

implementation of interaction protocols.

When communities become large (several hundred),

a backup centralized / decentralized knowledge is

desirable (in terms of reuse). Currently, we focus on

the construction of ontology and a consensus of own

CoPs.

REFERENCES

Bellifemine F., Poggi A., Rimassa G., 1999 «JADE, A

FIPA compliant agent framework”, Telecom Italia.

Boudriga, N. and Obaidat, M. 2004. Intelligent Agents on

the Web: A Review, Computing in Science and

Engineering, July/August, 35-42.

Boulanger, D., Dubois,G, “An Object Approach for

Information System Cooperation”. Inf. Syst. 23(6):

383- 399, 1998.

Brown, J. S., & Duguid, P. 1998. « Organizing

knowledge”. California Management Review, 40(3),

90-111.

Cohendet, P., & F. Créplet et O. Dupouët. 2003. «

Innovation organisationnelle, communautés :

croyances collectives et culture d’entreprises », Revue

d’économie politique.

Gasser, L. (2001). « Perspectives on Organizations in

Multi-agent Systems”. In: Luck, M., Marik, V.,

Stepankova, O., Trappl, R. (Eds.): Multi-Agent

Systems and Applications. LNAI 2086, Springer-

Verlag, Berlin, (2001), 1-16.

Gutknecht, O., Ferber, J., Michel, F, 2001.“Integrating

tools and infrastructures for generic multi-agent

systems”. In: Proceedings of the fifth international

conference on Autonomous agents, AA 2001, ACM

Press (2001) 441-448

Hammond M., 2004. Virtual Knowledge Communities for

Distributed Knowledge Management: A Multi-Agent-

Based Approach using JADE. Institut für Algorithmen

und Kognitive Systeme Universität Karlsruhe (TH)

SS.

Huhn, M. Stephens. 1999. Multiagent systems and

Societies of Agents. In Weiss, G (1999). Multi-Agent

Systems. MIT Press.

KMIS 2010 - International Conference on Knowledge Management and Information Sharing

376

JADE, 2006. (Java Agent DEvelopment framework),

Available at: “http://jade.tilab.com/”, cited 5

November.

Kenfack, C., 2007. “Modeling Community of Practices

Using Intelligent Agents”, CSCWD Melbourne,

Australia.

Nonaka I., 1991. «The knowledge creating company »,

Harvard Business Review, November- December.

Rigaud E., 2003. Définition et opérationnalisation d'une

organisation virtuelle à base d'agents pour contribuer à

de meilleures pratiques de gestion des risques dans les

PME-PMI. Thèse Informatique, école des Mines de

Paris à Sophia Antipolis .

Wenger E., 1998. “Communities of practice: learning,

meaning and identity”. Cambridge University Press –

New York.

Wooldridge, M. Jennings. N. 1999. Software Engineering

with Agents. Pitfalls and Pratfalls, IEEE Internet

Computing 3(3), pp. 20-27.

Wooldridge, M., Jennings, N. R., Kinny, D., 2000. “The

Gaia Methodology for Agent-Orient Analysis and

Design”. Journal of Autonomous Agents and Multi-

Agent Systems 3(3), Kluwer, pp. 285-312.

JADE AGENT IS INTERMEDIATION SYSTEM (JAIS) FOR KNOWLEDGE EMERGENCE IN COMMUNITY OF

PRACTICES - Intelligent Information Systems, Best Practices & Communities of Practice, Cooperation, Communication,

Collaboration and Knowledge Sharing

377