Validating Procedural Knowledge in the Open Virtual Collaboration Environment

214 views
146 views

Published on

Gerhard Wickler on "Validating Procedural Knowledge in the Open Virtual Collaboration Environment" at ISCRAM 2013 in Baden-Baden.

10th International Conference on Information Systems for Crisis Response and Management
12-15 May 2013, Baden-Baden, Germany

Published in: Technology, Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
214
On SlideShare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Validating Procedural Knowledge in the Open Virtual Collaboration Environment

  1. 1. 1Inte llig e nt Syste m s Track@ ISCRAM20 1 3Validating Procedural Knowledge inValidating Procedural Knowledge inthe Open Virtual Collaborationthe Open Virtual CollaborationEnvironmentEnvironmentGerhard WicklerGerhard WicklerAIAI, University of Edinburgh, UKAIAI, University of Edinburgh, UKg.wickler@ed.ac.ukg.wickler@ed.ac.uk
  2. 2. 2Inte llig e nt Syste m s Track@ ISCRAM20 1 3Introduction and OverviewIntroduction and Overviewcrisis response (planning) is a collaborative effort– idea: use new media technologies to support task-centric collaboration (use procedural knowledge)– problem: development and validation of proceduresOverview– Procedural Knowledge and OpenVCE– OpenVCE Workflow– Validating Procedural Knowledge– Evaluation and Future Work– Conclusions
  3. 3. 3Inte llig e nt Syste m s Track@ ISCRAM20 1 3Procedural KnowledgeProcedural KnowledgeSOPs: manual describing courses of action– represent best-practise knowledge; authored by experts– mostly available in form of books; used forteaching andtrainingproblems (with use in emergencies):– access time: finding best procedure in (large) body of text takestime, when time is short– structure: procedure described in free text form; must read allto find specific information– updating: procedure changes overtime; old knowledge tends topersist, especially in peoplestructure: hierarchical tasknetworks
  4. 4. 4Inte llig e nt Syste m s Track@ ISCRAM20 1 3OpenVCEOpenVCEOpenVCE: supports collaboration in virtual spaces– website based on open source content managementsystem(Drupal + MediaWiki) forasynchronouscollaboration– 3Dvirtual space forsynchronous collaboration (SecondLife)– virtual collaboration protocol: procedure forusingOpenVCE
  5. 5. 5Inte llig e nt Syste m s Track@ ISCRAM20 1 3Using Procedural Knowledge inUsing Procedural Knowledge inOpenVCEOpenVCE(Procedural)KnowledgeEngineeringKnowledgeValidation(Automated)PlanningIntelligentDistributedTo-Do ListExperts Crisis TeamExpertiseFeed-backDecisionsActionsCollaborationLessonsLearned
  6. 6. 6Inte llig e nt Syste m s Track@ ISCRAM20 1 3Collaborative SOPCollaborative SOPDevelopmentDevelopmentidea: use wiki as tool forcollaborative SOPdevelopment– MediaWiki: open source, scalable, robust tool– SOP/<I-N-C-A> extension: supports structuredprocedural knowledgecreating an SOPon the wiki:– write the unstructured procedure as free text– divide into articles; one method/refinement perarticle– markup articles using extension tagsvalidating procedural knowledge:– automatic analysis of formal aspects– test forconsistency– revise procedural knowledge
  7. 7. 7Inte llig e nt Syste m s Track@ ISCRAM20 1 3Distributed Use of ProceduresDistributed Use of Proceduresexport (formal) procedural knowledgeto planning toolresult: hierarchical tasknetwork– integrated into OpenVCE website– linked to procedural knowledge in wiki– linked to terminology in wikiplan execution:– people execute tasks– use capability model of agents involvedto distribute tasks– linktasks to code forautomaticexecution (e.g. tasks related to 3Dvirtual world)
  8. 8. 8Inte llig e nt Syste m s Track@ ISCRAM20 1 3Planning Domain ValidationPlanning Domain Validationproblemformulation: planning domain + problemdefinesearch space; essential forefficient planningidea:– explicitly represent (redundant) domain features– automatically (and efficiently) extract same features– ensure consistency of the representationoverview (domain features):– static vs. fluent relations– domain types– reversible actions– inconsistent effectsPDDL: contains types (optional); otherfeatures notsupported
  9. 9. 9Inte llig e nt Syste m s Track@ ISCRAM20 1 3Static vs. Fluent RelationsStatic vs. Fluent Relationsexample: DWRdomain(:predicates:static (adjacent ?l1 ?l2 - location):fluent (at ?r - robot ?l - location)etc.(:action move:parameters (?r - robot ?from ?to - location):precondition (and(adjacent ?from ?to) (at ?r ?from)(not (occupied ?to))):effect (and(at ?r ?to) (not (occupied ?from))(occupied ?to) (not (at ?r ?from)) ))domain validation:– static relations: must not appearin effects– fluent relations: should appearsomewhere in effects
  10. 10. 10Inte llig e nt Syste m s Track@ ISCRAM20 1 3Domain Types: ExampleDomain Types: Exampleexample:(:predicates:static (adjacent ?l1 ?l2 - location):fluent (at ?r - robot ?l - location)etc.(:action move:parameters (?r - robot ?from ?to - location):precondition (and(adjacent ?from ?to) (at ?r ?from)(not (occupied ?to))):effect (and(at ?r ?to) (not (occupied ?from))(occupied ?to) (not (at ?r ?from)) ))domain validation:– derive type systemfromoperatorspecification– compare declared types to derived types
  11. 11. 11Inte llig e nt Syste m s Track@ ISCRAM20 1 3Reversible Actions: ExampleReversible Actions: Exampleexample: DWRdomain(:action move:parameters (?r - robot ?from ?to - location):precondition (and(adjacent ?from ?to) (at ?r ?from)(not (occupied ?to))):effect (and(at ?r ?to) (not (occupied ?from))(occupied ?to) (not (at ?r ?from)):reverses (move ?r ?to ?from) ))domain validation:– compare operatorpairs: test whetherone reversesthe effects of the other– compare to “reverses” declaration in definition
  12. 12. 12Inte llig e nt Syste m s Track@ ISCRAM20 1 3Inconsistent Effects: ExampleInconsistent Effects: Exampleexample: DWRdomain(:action move:parameters (?r - robot ?from ?to - location):precondition (and(adjacent ?from ?to) (at ?r ?from)(not (occupied ?to))):effect (and(at ?r ?to) (not (occupied ?from))(not (at ?r ?from)) (occupied ?to) ))domain validation:– identify potential inconsistencies in effects– add (implicit) inequalities to prevent instances
  13. 13. 13Inte llig e nt Syste m s Track@ ISCRAM20 1 3EvaluationEvaluationsmall numberof planning domains (fromIPC)– domains were authored independent formthis work– domains were authored by expertsfeature extraction algorithms were applied to alldomains– runs in negligible time– provides feature values forall features in all domainscheckconsistency (manually)– no feature values in domain specificationsresults:– automatically extracted features appearsensible– conceptual flaw in one of the domains was highlighted– also: minorissue in planner
  14. 14. 14Inte llig e nt Syste m s Track@ ISCRAM20 1 3ConclusionsConclusionsOpenVCE:– wiki extension supports collaborative development andvalidation of (semi-formal) procedural knowledge– virtual collaboration supported by procedural knowledgefeatures:– static vs. dynamic: trivial, but helpful– types:» derived type system is most specific (of its kind)» flat; not hierarchical taxonomy/ontology– inconsistent effects: useful forplanning– reversible actions: necessary criterion only
  15. 15. 15Inte llig e nt Syste m s Track@ ISCRAM20 1 3The FutureThe FutureHedlamp: Machine Learning and Adaptation ofDomain Models to Support Real Time Planning inAutonomous Systems– automatically acquiring procedural knowledge (machinelearning)– domain analysis: towards a more human-likeunderstanding of procedural knowledge (used inplanning context)– use domain analysis to automatically improve learnedmodelsee: http://www.aiai.ed.ac.uk/project/hedlamp/orgo there: http://virtual.aiai.ed.ac.uk:8002/Hedland/– use in Firestorm OpenSimbrowser

×