Abstract | ||
---|---|---|
The k-means method is a simple and fast clustering technique that exhibits the problem of specifying the optimal number of clusters preliminarily. We address the problem of cluster number selection by using a k-means approach that exploits local changes of internal validity indices to split or merge clusters. Our split and merge k-means issues criterion functions to select clusters to be split or merged and fitness assessments on cluster structure changes. Experiments on standard test data sets show that this approach selects an accurate number of clusters with reasonable runtime and accuracy. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/DEXA.2009.39 | DEXA Workshops |
Keywords | Field | DocType |
clusters preliminarily,k-means method,automatic cluster number selection,clustering technique,cluster number selection,split and merge,cluster structure change,-k-means,accurate number,k-means approach,merge k-means approach,cluster number selec- tion,optimal number,validity indices,k-means issue,criterion function,k means,structural change,merging,clustering algorithms,statistical analysis,indexes,data mining,silicon,frequency modulation | Data mining,k-means clustering,Cluster (physics),Pattern clustering,Computer science,Determining the number of clusters in a data set,Test data,Cluster analysis,Merge (version control),Statistical analysis | Conference |
Citations | PageRank | References |
10 | 0.60 | 8 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Markus Muhr | 1 | 74 | 5.53 |
Michael Granitzer | 2 | 822 | 80.14 |