DSDBSCAN: A Novel Clustering Algorithm Based on Double Sampling for DBSCAN

Download Now
Provided by: Binary Information Press
Topic: Data Management
Format: PDF
DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based spatial clustering algorithm, which can discover clusters of any arbitrary shapes and handle noises effectively. But when the DBSCAN algorithm is used to cluster and analyze large-scale spatial databases, it requires large volumes of memory support and I/O cost. Aimed at the shortage of DBSCAN, on the one hand, the sampling method is taken into consideration to diminish the data size without affecting characteristics of the whole data set, at the same time, in order to lower the randomicity produced by data sampling, an interfering rank function is introduced to assist the sampling process, as the testing results prove that this kind of sampling creates well-proportioned data set.
Download Now

Find By Topic