Dynamic Agent Communities Facilitating to Distant Learning in a Virtua - PDF document

1 1 Dynamic Agent Communities Facilitating
1 Dynamic Agent Communities Facilitating to Distant Learning in a Virtual University Information SpaceVadim Ermolayev Dept. of Mathematical Modelling and Information Technologies,ZaporozhyeState University,66, Zhukovskogo st., 330600, Zaporozhye, Ukraine,E-mail: eva@zsu.zaporizhzhe.ua EVA: IS2000-VUDE, 06.11.20002Motivation and Background Layered mediator University IS with unified interfacesHierarchy of distributed wrapped legacy IS and information resourcesI guess I know how to use this!!! It’s like doing my everyday work, but electronically!!!VUISVUIS is considered to be a passive media inhabited by active “beings” (agents) representing real life actors… It’s like CA Real World or Microsoft Encarta 2 EVA: IS2000-VUDE, 06.11.20003Active VUIS InhabitantsPersistent MAS and Member Agents SUR[\ UR[\ Generic MAS agents Middle Agent Middle Agent Ontology Agent SDS Coordination Agent CloningAgentUniversity MAS Pr

2 oxy Agent(s)Lower level V Dept. MASProxi
oxy Agent(s)Lower level V Dept. MASProxies respective MAS and are the representative members in the higher level MAS EVA: IS2000-VUDE, 06.11.20004Agents’ Interactions –Parametric Feedbacks i j ),(YXfa
0.2 0.4 0.6 0.8 15 No of students at the classProfessors interest in new students )(~1Xy Entry test difficulty levelParametric feedbacks expressed attitudecapability(policy) at a certain state–accept a new student to the class 3 EVA: IS2000-VUDE, 06.11.20005Agent Communitiesas dynamic coalitions for performing tasks (sets of works)there are MAS already-tasks emerge and are soon accomplished -agents join the coalition if they are capable to contribute to the task execution-an agent may participate in more than one coalitions at a timeHow?accept or reject the work, react with parametric feedbacksWhat is changing in time?-the coalitionagents join and move out-agentcapability to perform a work -agents knowledge(belief

3 s) about the capabilities EVA: IS2000-V
s) about the capabilities EVA: IS2000-VUDE, 06.11.20006Framework Models Generic Agent ModelCommunication ModelCoordination ModelFunctional System / Component ModelProcess Model???? Evolution Model ???? 4 EVA: IS2000-VUDE, 06.11.20007Agents proactive self-development and self-adaptation in response to the changes in MAS and environment caused by task(s) executionAgents Evolution ModelCapabilitiesBeliefs 1 n ijnwwAjcXccjji( jw ,0 i j ),(~XsfYi -parameter constraints-policy constraints-transition function EVA: IS2000-VUDE, 06.11.20008Case Study VUDE Domain Professor(PRAAssistant(), Course master(CMALibrarian () agents External influences andMASreactions DepartmentMAS Middle Agent Task communities Influences Facilitator cloning Ontology queries Co-ordination Middle Agent Middle Agent Tutor Agent(User) Tutor Agent(User) Ontology Agent Proxy Agent SDS Coordination Agent CloningAgent Proxy Agent(

4 secretary)PhD recruiting scenario:Phase
secretary)PhD recruiting scenario:Phase 1:A PhD candidate submits the CV and indicates his/her intention to become a PhD student (Phase 2:The CV is analysed and the best Professor Match is searched (Phase 3:Qualified candidate passes the test from the chosen professor(Phase 4:Successful candidate is interviewed and assigned to a research project (Phase 5:The professor and his assistant prepare the individual curriculum for the accepted candidate as well as the list of recommended reading (CMA 5 EVA: IS2000-VUDE, 06.11.20009 VDept Agents and their Roles(w.r.t. the Case Scenario) PA Zaporozhye State UniversityInst. of Mathematics and CyberneticsElena VetukhVyachyslav TolokSer ey BorueVadim ErmolayevSer ey GomenyukVitaly MukhinNadezhda MatveyshinaNatalya KeberleVladimir DavidovskyMaxim ShumakovVictor GristchakAnatoly DoroshenkoSecretary Mrs Chair Prof. PR1 Prof.

5 PR2 Prof.
PR2 Prof. PR3 Ass. A1 Ass. Prof A2 Ass. Prof A3 Ass. Prof A4 Ass. Prof A5 Ass. Prof A6 Visiting: Prof. PR4Virtual: Prof. PR5 Faculty:ProfDept. Of Mathematical Modelling and ITProxy Virtual Faculty A2 A3 A4 A5 A6 PR2 PR3 PR4 PRA5 PRA1 AA1 CMA PRA1 Utility Agents:Clone a TA each time a new PhD candidate comes Monitor agents activities, Accept work results to its SDS,Provide work params on request.Provide common dept. ontologies on request Middle Agents:PRA1-PRA5 evaluate candidates qualifications; provide a test, evaluate test results, arrange the interview, provide course recommendationsprepare working plan, prepare curriculumCMAprovide electronic courses, issue calls for new coursesprovi

6 de electronic textbooks, issue calls for
de electronic textbooks, issue calls for new teaching materials Proxy Agent: PAaccept outer influences, order to clone , pipeline the candidate to his EVA: IS2000-VUDE, 06.11.200010Phase 3. Testing, Step 1 Test is ordered.Modelling( How did they play their roles? nt CoA No Results t execute },,(111YXt
,'_22YX
,'33YX
,'44YX
,'_55YX
No works for redirectionRequire_Testpolicy:1.Generate4 works2.InvokeTAto influence with itself with Just monitoring… 6 EVA: IS2000-VUDE, 06.11.200011 3ww PRA CoA Results: Phase 3. Testing, Step 2 Test is provided.Modelling( How did they play their roles?execute g delayexecute, delayMonitoring and accepting feedback(s)redirect to itself TA redirect to itself Provide_Testpolicy:1.Generatetest file ( )according to the ontology and received params 2.InvokePRAto influence with: Accept_Results, ) EVA: IS2000-VUDE, 06.11.200012Phase 3. Testing, Step 3 Test is performed.Model

7 ling( How did they play their roles? 3ww
ling( How did they play their roles? 3ww PRA CoA Results: execute delayexecute, delayMonitoring, accepting feedback(s), providing work parametersredirect to itself TA redirect to itself execute, delay PRA Params: Perform_Testpolicy:1.Require and acceptparameters from 2.Invokehuman user to fill in the test form. 3.InvokeTAto influence with: Accept_Results, ) delay 7 EVA: IS2000-VUDE, 06.11.200013Phase 3. Testing, Step 4 Test results are evaluated.Modelling( How did they play their roles? w PRA CoA Results: execute delayexecute,delayMonitoring, accepting feedback(s), providing work parameters redirect to itself execute,delay PRAParams: delay delay PRAexecute EVA: IS2000-VUDE, 06.11.200014Phase 3. Testing, Step 5 Parametric Marks are analysed.Modelling( How did they play their roles? PRA CoA execute delayexecute,delayMonitoring, providing work parameters Negative:inform execute,delay PRAParams: delay delay

8 PRAexecute 5w execute g Positive:Interv
PRAexecute 5w execute g Positive:Interview Phase 4.No Results 8 EVA: IS2000-VUDE, 06.11.200015SummaryWhat has been done:What has been not, yet:The processes of DE may be modelled and executedas tasks without pre-defined task plansin frame of the formal agent-based approach to modelling and management of the processes of information interchangeThe particularities of the framework generic agentskeletons, parametric feedbacks-agents ability to evolveprovide for the scalabilityand pro-active dynamic characterof the approach. Planned for short-term:Architectural frameworkon the base of the presented formal approachSemantic issues:Ontology representation, interoperability, knowledge sharing among generic agents, within MAS and task communities.Planned for middle-term:-Design and implementation of the prototypefor a Virtual University Department-Further evaluationof the applicability of the approach to VUDE, BPM, DLIB dom