Section 7.1ObjectivesUnsupervised ClassificationK-means ClusteringAssignmentReassignmentEuclidean DistanceManhattan DistanceForced ClusteringDemonstrationSection 7.2Slide 12PowerPoint PresentationSlide 14Slide 15Slide 16DemonstrationSelf-Organizing Map ResultsSection 7.1K-Means Cluster Analysis2ObjectivesDiscuss the concept of k-means clustering.Define measures of distance in cluster analysis.Understand the dangers of forced clustering.Generate a cluster analysis and interpret the results.3Unsupervised Classificationcase 1: inputs, ? case 2: inputs, ? case 3: inputs, ? case 4: inputs, ? case 5: inputs, ? Training Datanew casenew casecase 1: inputs, cluster 1 case 2: inputs, cluster 3 case 3: inputs, cluster 2 case 4: inputs, cluster 1 case 5: inputs, cluster 2Training Data4K-means Clustering5Assignment6Reassignment7Euclidean Distance(U1,V1)(U2,V2)L2 = ((U1 - U2)2 + (V1 - V2)2)1/28Manhattan Distance(U1,V1)(U2,V2)L1 = |U1 - U2| + |V1 - V2|9Forced ClusteringLast MonthNext Month100% Cotton60/40 Blend 60/40 Blend6236232850285010This demonstration illustrates conducting cluster analysis with Enterprise Miner.DemonstrationSection 7.2Self-Organizing Maps12ObjectivesDiscuss the concept of self-organizing maps.Generate a self-organizing map and interpret the results.13Self-Organizing MapsWinner!NeighborNeighborNeighbor14Winner!NeighborNeighborNeighborSelf-Organizing Maps15Self-Organizing Maps16Self-Organizing Maps18This demonstration illustrates generating a self-organizing map with Enterprise Miner.Demonstration19Self-Organizing Map Resultsyounger, unmarried malesunmarried malesmarried malesyounger, unmarried femalesunmarried femalesmarried