OUTLINE CAPITALIZATION OF COLLECTIVE KNOWLEDGE: Knowledge - PDF document

OUTLINE CAPITALIZATION OF COLLECTIVE KNOWLEDGE:  Knowledge management and Knowledge Engineering FROM KNOWLEDGE  Definitions ENGINEERING, MULTI-  Process AGENTS TO CSCW AND SOCIO  Knowledge Capitalization approaches  Cooperative Knowledge SEMANTIC WEB  Defintinions  Traceability and capitalization approaches  Socio-semantic Web  Definition  Semantic Web 1 Nada Matta 1 and Davy Monticolo 2  Examples 1Tech-CICO-University of Technology of Troyes nada.matta@utt.fr 2 SET-University of Technology of Belfort Montbeliart davy.monticolo@utbm.fr 2 KNOWLEDGE  Knowledge is data, information used in a given context  Knowledge is [Polyani] KNOWLEDGE MANAGEMENT AND  Tacit KNOWLEDGE ENGINENRING  Explicit 3 4

KNOWLEDGE MANAGEMENT CYCLE OUTLINE [NONAKA & TAKEUSHI]  Knowledge management process ICT Semantic Web Combination  Knowledge Enginnering process  Corporate Memory Explicit Explicit  Knowledge Engineering approaches  CommonKADS Internalization Externalization  MASK Tacit Tacit Knowledge Engineering Socialization Community of 5 6 Practices KNOWLEDGE MANAGEMENT CYCLE KNOWLEDGE ENGINEERING [NONAKA & TAKEUSHI] Knowledge engineering is an approach allowing problem solving extracting and modelling [Aussenac, Combination Corporate Memory Bradshow], [Newell] • What Explicit Explicit Conceptual • Why Model • How Internalization Externalization TextMining Interviews Tacit Tacit Observation … Knowledge KB Documents Engineering Socialization Experts 7 8

PROFESSION MEMORY CORPORATE MEMORY Profession memory is the externalization of the knowledge produced in and for a given domain « A corporate memory is a persistent and explicit  Structure : representation of knowledge and information of  Definition of the problem (or the process) an organization » [Van Heijst, 96], [Dieng et al, 03]  Problem solving methods  Description of manipulated concepts Several memory types: Profession memory, project memory, management memory 9 10 KNOWLEDGE ENGINEERING COMMONKADS [BREUKER ET AL,94] APPROACHES  CommonKADS  Generic models Generic Models  MASK Library  Process Adaptation  Models MC Selection Formalisation, Implementation Analysis 11 12

TASK TYPES GENERIC MODELS LIBRARY Task Types  Generic task models Analysis Modification Synthesis  Problem solving methods library [Benjamins]. Classify Prediction Repare Change Design Planning Diagnostic Evaluate Monitoring 13 14 Prediction Diagnostic Heuristics Abnormal Observations Observations Generate Symptom Behaviour Specify hypotheses Environment Observations detection description System Normal Observations Transform Discriminate Diagnostic Hypotheses hypotheses New 15 16 behaviour Additional Observations

Monitoring Evaluate Environment Select Observations Transform Variable Values Case Elements Abstract Description Differences Compare Decision Match Measures History classes Parameter Classify Values System Specify Norms Model System Instantiate Parameters Select model Difference Classes 17 18 Criteria Design Planning Plan models Specify Identify Goal list Intention Environment Specify Objective Violated Requirements Evaluate Tasks, Time, ... requirements Composites Build Artefact Rules/Laws No satisfied Evaluate Assemble Plan Objective 19 20

THE MASK METHOD [ERMINE, CML : CONCEPTUAL MODELLING MATTA, CASTILLO] LANGUAGE Level Entity Consensus Problem solving T ask, Task Task structure Knowledge Book Inference, Inference Sensibilisation Inference structure Training Co-building Domain Concept, relation, Domain expression, attribute 21 22 PROCESS MODEL PB SOLVING MODEL Verification product stitch Knowledge programming Tools Knowledge on Electronic machines machine 2 Verify the product in Verify the good “fontures” Knowledge Look at the weak places Sketch Conceive the shape flattering position (smooth goes well with - styling and model fallen from machine Shape table with or without light) model or by wearing making test Model - technique on the Realization chosen stitches Technician of the product Prototype Verify the Verify all the dimension in Verify the weak sensitive points height, width, 1/2 places: Edges Motives scale Technician Armholes (crossing sleeve / body) Decreases Validated 23 24 Technician program

REFERENCES CONCEPT MODEL Knowledge Acquistion as modelling, M. Ford, J.M. Bradshaw (Eds), 1993  CommonKADS Library for expertise modelling, Reusable problem solving components, J. Breuker  and W. Van de Velde (Eds), IOS press, Amsterdam, 1994 A. Newell, The Knowledge level, Artificial Intelligence Journal, 19(2), 1982  Edition [Van Heijst et al, 97] Van Heijst G., Schreiber A. Wielinga B., Using Explicit Ontologies in KBS  Development. International Journal of Human Computer Studies , Vol. 46, 1997. Breuker J., Van de VeldeW., Common-KADS Library for expertise modelling Reusable problem  solving components, Frontiers in Artificial Intelligence and Applications , J. Breuker and W. Van de Velde (EDS), Amsterdam: IOS.Press 1994. Edition for the structures Benjamins R., Problem solving methods of diagnosis , Rapport de Thèse de l'université d'Amsterdam, Minimal edition  ISBN 90-9005877-X, Amsterdam, 1993. Acquisition et Ingénierie des connaissances, Tendances Actuelles, Coordination: N. Aussenac-Gilles,  P. Laublet, C. Reynaud, Cepadues Editions, 1996 [Nonaka et al, 95] Nonaka I., Takeuchi H.: The knowledge-Creating Company: How Japanese  Abolition of Companies Create the Dynamics of Innovation . Oxford University Press, 1995 Reduction Crossing Edge rib Main Necklace Reverse the Ermine J.L. – Les systèmes de connaissances, Eds. Hermès 1996 (2nd Edition 2000)  edition sleeve auxiliary Méthodes et Outils pour la gestion des connaissances, R. Dieng, O. Corby, A. Giboin, J.  edition Golebiowska, N. Matta, M. Ribière, Dunod, 2000 Dieng-Kuntz R., Matta N., Knowledge Management and Organizational Memories, Kluwer  Academic Publishers, 2002. Matta N. Zaher L., Applications of knowledge Engineering for Design, Methods and Tools for  effective Knowledge Life-Cycle Management , Bernard A., Tichkiewitch S. (Eds), Springer, 2008. CASTILLO, O., et MATTA, N. A knowledge acquisition system for the French Textile and Apparel.  Knowledge-Based Intelligent Information and Engineering Systems: 9th International Conference, 25 26 KES 2005, 14-16 September 2005, Melbourne. 6 p. CASTILLO, O., et MATTA, N. An operational learning system definition. 19th International Joint  Conference on Artificial Intelligence IJCAI 2005 – Workshop on knowledge Management and Organizational Memories, 1-5 August 2005, Edimbourg. 12 p. Polyani M. (1958), Personal knowledge , Chicago : University of Chicago Press, 1958.  OUTLINE  Cooperative Knowledge definition  Project memory COOPERATIVE  Traceability approaches KNOWLEDGE  Traceability and capitalization approache 27 28

PROJECT MEMORY Explicitation of the experience learned during project realization [Matta] COOPERATIVE KNOWLEDGE Process déc 4 obj 4 cc q 3 q 4  Knowledge produced in cooperative activity a c4 déc 5 a d5 obj 5 (projects, cooperative decision making, etc.): r hu4 , r ma4 , r in4 r hu5 vd 5 r ma5 co 5 r in5 cr  Organizational dimension: actors, tasks, resources, 5 q 5 constraints,  Cooperative dimension: negotiation, argumentation, Design Rationale Product representation Problème etc. compresseur Ca/ Ext Proposition Proposition Vue Technologue Ca/ Ar Carter Carter Cart/ Arbre Guider en rotation l’arbre/carter Argument Argument Argument Arbre Guider en rotation Ar /Ext l’arbre/carter Arbre Ar/Ca Arbre /Cart Decision Goals Constraints Techniques Requirements Directives Tools Procedures 29 Methods Competences 30 Relationships References Roles Environment and organization Approaches to handle cooperative TRACEABILY METHODS knowledge Representation of the dynamics Representation guided by of problems solving the decision-making  CSCW-Design rationale: Constraint  IBIS [Coklin, 98], QOC [McLean, 91] , DRAMA [Brice] Problem Problem Interpretation (Design rationale tree) Task Proposition Proposition Proposition  DIPA (Problem solving model) [Lewkowicz, 99]  DRCS (Graphs : Concepts, Relations) [Klein, 93] Decision Argument Argument Argument Argument  Project Management: Artifact  DRCS (Graphs : Concepts, Relations) [Klein, 93] IBIS, QOC, DRAMA DIPA, DRCS 31 32

QOC: QUESTION, OPTION, CRITERIA TREE TRACEABILY AND CAPITALIZATION [MCLEAN, 91] METHOD Flux Numerical Decision Making Organization Connexion type ? Hybrid Performance Product Analogical Cost DYPKM [Bekhti, Djaiz, Matta] 33 34 Support Objection QOC: OPTION LINKS DIPA [LEWKOWICZ, 00] Flux  Cooperative problem solving model Numerical Connexion type ?  Structure based on principal concepts Performance Hybrid Analogical Cost Performance Optical connexion Installation 35 36