Abstract | ||
---|---|---|
Multi Density DBSCAN (Density Based Spatial Clustering of Application with Noise) is an excellent density-based clustering algorithm, which extends DBSCAN algorithm so as to be able to discover the different densities clusters, and retains the advantage of separating noise and finding arbitrary shape clusters. But, because of great memory demand and low calculation efficiency, Multi Density DBSCAN can't deal with large database. Therefore, GCMDDBSCAN is proposed in this paper, and within it 'migration-coefficient' conception is introduced firstly. In GCMDDBSCAN, with the grid technique, the optimization effect of contribution and migration-coefficient, and the efficient SP-tree query index, the runtime is reduced a lot, and the capability of clustering large database is obviously enhanced, at the same time, the accuracy of clustering result is not degraded. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/DASC.2013.115 | 2013 IEEE 11TH INTERNATIONAL CONFERENCE ON DEPENDABLE, AUTONOMIC AND SECURE COMPUTING (DASC) |
Keywords | Field | DocType |
GCMDDBSCAN, DBSCAN, multi-density, grid, contribution, migration-coefficient | OPTICS algorithm,Data mining,Clustering high-dimensional data,Correlation clustering,Computer science,Determining the number of clusters in a data set,SUBCLU,Cluster analysis,DBSCAN,Grid | Conference |
Citations | PageRank | References |
1 | 0.35 | 3 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Linmeng Zhang | 1 | 1 | 0.35 |
Zhigao Xu | 2 | 2 | 1.73 |
Fengqi Si | 3 | 3 | 3.45 |