Dynamic and Incremental Clustering Based on Density Reachable
Yu-Chen Song, Hai-Dong Meng, Shuling Wang, Michael J. O’Grady, Gregory M. P. O’Hare · 2009
The traditional clustering algorithms are only suitable for the static datasets. As for the dynamic and incremental datasets, the clustering results will become unreliable after data updates, and also it will certainly decrease efficiency and waste computing resources to cluster all of the data again. To overcome these problems, a new incremental clustering algorithm is proposed on the basis of density and density-reachable. Theoretical analysis and experimental results demonstrate that the incremental algorithm can improve the efficiency of data resource utilization, and handle the dynamic datasets effectively.