Publication | Closed Access
Automatic Method for Determining Cluster Number Based on Silhouette Coefficient
114
Citations
1
References
2014
Year
Cluster ComputingEngineeringMeaningful GroupsBiometricsSilhouette CoefficientCombinatorial Data AnalysisUnsupervised Machine LearningCluster TechnologyData ScienceData MiningPattern RecognitionBiostatisticsComputational GeometryStatisticsDocument ClusteringClustering (Nuclear Physics)Knowledge DiscoveryData PatternsComputer ScienceCluster DevelopmentNatural SciencesClustering (Data Mining)Fuzzy Clustering
Clustering is an important technology that can divide data patterns into meaningful groups, but the number of groups is difficult to be determined. This paper proposes an automatic approach, which can determine the number of groups using silhouette coefficient and the sum of the squared error.The experiment conducted shows that the proposed approach can generally find the optimum number of clusters, and can cluster the data patterns effectively.
| Year | Citations | |
|---|---|---|
Page 1
Page 1