CBR for DesignDesign ResearchReuse of DesignDesign TaskDesign Task (figure)Approaches for design tasksThe PCM ModelSlide 8Mapping Design Task to CBR-cycleCase Based DesignA Framework for CBD SystemsCharacteristics of CBD SystemCBR System ArchitectureCase Retrieval for Innovative/Creative designCase RetrievalFish & Shrink AlgorithmFish & ShrinkSlide 18Fish & Shrink MethodSlide 20Fish and ShrinkStructural Similarity Assessment and Adaptation Using GraphsRequired FunctionalityStructural similarity assessment and adaptationSlide 25Example of structural similarity assessment: TOPOCombination GraphSlide 28Structural Adaptation by Case CombinationSlide 30SummaryReferencesCBR for DesignUpmanyu MisraCSE 495Design ResearchDevelop tools to aid human designersAutomate design tasksBetter understanding of designIncrease quality Take lesser timeImprove predictability of design REUSEReuse of DesignReuse old design – share intellectual property (IP)As the ‘reuse’ increases, the complexity increasesHuman assistance is mandatoryDirected towards assisting human designer rather than making intelligent decisions by ownDesign TaskRoutine design-is completely a part of a set of potential designs-all variables, their ranges, and knowledge to compute their values are directly derivable from the set-easily implementedInnovative Design-is partially derived from the set of potential designs-all components need to be derived. The knowledge is incomplete-design needs to be iteratively derivedCreative Design-no overlap with the set of potential design. The set needs to be extended-All components need to be definedDesign Task (figure)Approaches for design tasksFormulaeConstraintsRules and grammarsCBRPrototype based reasoningRoutine Only- Goel (1989), Domeshek and Kolodner (1992), Hinrichs (1992)The PCM ModelPropose – involves using domain knowledge to map part or all of the specification to partial or complete design proposalsCritique – assessment of the proposed design solution Modify – takes info about a failure of a proposed design as its input and then changes the design to get closer to the desired specificationThe PCM ModelThe CBR CycleMapping Design Task to CBR-cycleCase Based DesignDefined as “the process of creating a new design solution by combining and/or adapting previous design solution(s)”useful tool for intelligent system design in a domain where either an explicit model does not exist or one is not yet adequately understoodcan learn from interaction with userA Framework for CBD SystemsCharacteristics of CBD SystemCan produce complete and complex designs based on relatively small knowledge basedesign starts from complete cases, implicitly achieving trade-offs between several constraintsdesign history of existing cases makes design problem solving more efficientusing cases as a source of knowledge allows learning by storing new casesCBR System ArchitectureFour Knowledge Containers-Vocabulary: should be able to capture all salient features of the design. Task dependent-Case base: -- usage: cases can capture both regular/normal situations as well as exceptions/abnormal situations-- granularity: for task-oriented user support, the grain size of the cases matches that of the decisions made-Similarity measures: to compare queries and cases in their corresponding representations-Solution transformation: contains knowledge required to evaluate solutionsCase Retrieval for Innovative/Creative designFlexible case retrievalStructural similarity assessmentSimilarity assessment in terms of adaptabilityCase RetrievalFlexible Case Retrieval – Given a large case base, a problem, and a number of aspects that are relevant for similarity assessment, a set of cases is to be retrieved which show similar aspects as in the actual problemThe aim is to exploit different views on single casesImportance of certain aspects for similarity assessment may not be known at memory organization time- dynamic weighting is required - use kd trees, Case Retrieval Nets etc.Fish & Shrink AlgorithmUsed for Flexible Case RetrievalSelects and ranks potential cases from a large set of casesConsiders different aspects (representations) of casesMain idea “it should be more efficient to avoid searching in the nearby neighborhood of cases which have already been found to be inappropriate”Fish & ShrinkA representation function takes the case and outputs the aspect in the desired representation space : case 1: (20, empty, 0.05) case 2: (19, half-empty, 0.9)A distance function that can take two representations in space and calculate the distance of the two cases in this aspect namenamenameemptyhalfcas e )2(220)1(1casename2(case1,case2) 0.5emptycase )1(219)2(1caseFish & ShrinkFish & Shrink MethodView distance SDThe view distance from the query to some case is called test distance, and the view distance between two cases is called the base distanceThis is a basic distance function, researchers generally use their ownPresumption: View distance function have to satisfy the triangle equalitySD(Fx,Fy,W ) wii(Fx,Fy)Fish & Shrink AlgorithmFish and Shrink01Distance to the queryT1T2T3Structural Similarity Assessment and Adaptation Using GraphsTo retrieve structurally most similar casesStructured case representation → GraphFind maximum common sub graphCAD example for industrial building:Object represented by set of attributes describing its geometry and typeDifferent pipe system shows different topological relationsBuilding structure can be mapped onto its pipe systemRequired Functionality A compile function is used to translate the selected attributes and their relations into graphsA recompile function is used to translate the selected solution graphs into their attributes-based representationRetrieving case is conducted by selecting the case having maximum common sub graph with the problemStructural adaptation proceeds by combining case parts that are not included in the sub graphStructural similarity assessment and adaptationA graph g=(V,E), where V is the set of vertex andmcs(G) is the maximum common sub graph of a set of graphs GLet be the set of all graphs, O be the a finite set of objects represented by attribute values
View Full Document