Slide 1Slide 2Slide 3Slide 4Bayesian NetworkSlide 6Framing, Constructing and Assessing Bayesian ModelSlide 8Slide 9Slide 10Slide 11Slide 12Lumière/Excel in OperationLumière /Excel in OperationBeyond Real-Time AssistanceSlide 16Ongoing Work and SummarySummaryKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligenceMonday, March 13, 2000Yuhui LIUDepartment of Computing and Information Sciences, KSUReadings:“The Lumière Project: Bayesian User Modeling for Inferring the Goals and Needs of Software Users ”- Horvitz, Breese, Heckerman, Hovel and RommelseLecture 24Lecture 24 Uncertain Reasoning Presentation (3 of 4): Decision Support Systems and Bayesian User ModelingKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligencePresentation OutlinePresentation Outline•Goal–Bayesian User Model used in reasoning under uncertainty to capture the relationships among user needs, user actions, and user query•Structure–Background knowledge of Bayesian User Model–Some difficulties in Lumière project implementation–Introduction of Lumière/Excel prototype–Office assistant-- Lumière/Excel prototype in real world•References:–Machine Learning, T. M. Mitchell–Artificial Intelligence: A Modern Approach, S. J. Russell, and P.Norvig–Trouble Shooting under Uncertainty, David Heckerman, John S. Breese, and Koos Rommelse–A Tutorial on Learning With Bayesian Networks, David Heckerman–Lecture Notes in CIS 798, William HsuKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligencePresentation OutlinePresentation Outline•Outline–Background: Bayesian User Models–Lumière Project Implementation•Structuring Bayesian User Models•Temporal Reasoning about User Action•Bridging the System Events and Users Actions –Lumière/Excel System Prototype–Lumière in Real World--Microsoft Office Assistant –Future Work and Summary-Issues–How to build an appropriate Bayesian User Model?–How to fulfill temporal reasoning?–How to connect system event to user actions?–Is Lumiere/Excel prototype applicable to real world software application?Kansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial Intelligence•A Graphical probabilistic model combining Bayesian Network and influence diagrams makes inference about the goals of users •Features–Express uncertainty–Incorporate prior knowledge–Support decision making–Be able to reason over time–Provide a decision theoretic model and provide utility values for the decision nodes with influence diagram •General Product Rule in this model: Background: Bayesian User ModelBackground: Bayesian User Model    niiinXparentsXPXXXXP1321,......,,Kansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligenceBayesian NetworkBayesian Network•Example 1:BatteryAgeBatteryEngineTurns OverStarterFuel PumpFuel SubsystemEngine StartsSpark PlugsFuel LineFuelFuelGaugeLightsKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligenceBayesian NetworkBayesian Network•Example 2:FraudAgeSexGasJewelryP(f=yes)=0.00001P(a=<30)=0.25P(a=30-50)=0.40P(s=male)=0.5P(g=yes|f=yes)=0.2P(g=yes|f=no)=0.01P(j=yes|f=yes,a=*,s=*)=0.05P(j=yes|f=no,a=30-50,s=male)=0.0004 P(j=yes|f=no,a=>50,s=male)=0.0002 P(j=yes|f=no,a=<30,s=female)=0.0005 P(j=yes|f=no,a=30-50,s=female)=0.002P(j=yes|f=no,a=>50,s=female)=0.001Kansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligenceFraming, Constructing and Assessing Framing, Constructing and Assessing Bayesian ModelBayesian Model•Several important evidential distinctions–Search–Focus of attention–Introspection–Undesired effects–Inefficient command sequences–Domain-specific syntactic and semantic content•A Small Bayesian Network in Lumiére projectDifficulty of current taskUser of expertiseUser needs assistanceUser distractedRecent menusurfingPause after activityKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial Intelligence•Markov Model–Dependencies among variablesat adjacent time periods. •Time-Dependent Probability Approach –Alternative goals at thepresent moment–Temporal model-construction methodology–Less relevance of earlier observation to the current goals–Definition of evidential horizon and decay parameters Temporal reasoning about user Temporal reasoning about user actionsactions 1Kansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligenceSystem events and Users actionsSystem events and Users actions Time stamped atomic events Modeled events Lumière events architecture:Lumière Events LanguageExample primitives:Rate(xi,t), Oneof({x1,…….xn},t), All({x1,…….xn},t), Seq(x1,…….xn,t), TightSeq (x1,…….xn,t), Dwell(t)Build and modify transformation function which be compiled into run_time filter for modeled eventsKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial Intelligence•Overall Lumière/Excel ArchitectureLumére/Excel ProjectLumére/Excel ProjectKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial IntelligenceLumière/Excel ProjectLumière/Excel Project• Control policies of timing for assistance–Pulsed strategy–Event-driven control policy–Augmented pulsed approach–Deferred analysis• User profile –Tailor Lumiere/Excel performance according to user’s expertise. –Update the probability distribution over the user’s needs. –Determine special competency variable which can be used to estimate the expertise in Bayesian user modelKansas State UniversityDepartment of Computing and Information SciencesCIS 830: Advanced Topics in Artificial Intelligence•Lumière’s instrumentationLumière /Excel in OperationLumière /Excel in OperationStreams of eventsProbability distribution over user’s needsStream of Atomic Events and ObservationsProbabilities Distribution of Inferred NeedsLikelihood of Needing AssistanceKansas State